.?
;. w! 2.;
. «wa. . c .. J .
‘. 53w", ..
xfifiuaw H; u;

y .
2

. J.

,2

Emma“. g

:4“. .._.
:5; 3. ,
. I. '4’ ‘
. £1. ”MEL
‘ Muhmwwmmm...

~
5 v
«I S

3”. 2:.
. EH Emu r.»
mummy. a .
.u. .véwfin

F:

r

chm...

r: s

. wfififi .
..

.rflukfikfi ...

“n.3,;
5., man

and»...

Jr ‘ .5
.2...
H!

5a.

 

TH E St S
7

7501

This is to certify that the
dissertation entitled
THE RESEARCH PRODUCTIVITY OF FAMILY MEDICINE
DEPARTMENT FACULTY: A NATIONAL STUDY

presented by

JOSEPH J . BROCATO
has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Dept. of Educational
Administration

 

 

//7/ %

Major professor

Date z/s/O/

MSU i: an Affirmative Action/Equal Opportunity Institution 042771

 

 

LIBRARY

Michigan State
UnIverslty

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

8% I g goes;

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClFlC/DateDue.p65-p.15

 

THE RESEARCH PRODUCTIVITY OF FAMILY MEDICINE DEPARTMENT
FACULTY: A NATIONAL STUDY
By

Joseph J. Brocato

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Educational Administration

2001

ABSTRACT
THE RESEARCH PRODUCTIVITY OF FAMILY MEDICINE DEPARTMENT
FACULTY: A NATIONAL STUDY
By

Joseph J. Brocato

The purpose of this study was to determine the research productivity of US. medical
school family medicine department faculty and the individual faculty characteristics,
prior socializing experiences, and research environment characteristics predictive of their
research productivity. A national survey was sent to random sample of 796 faculty
holding the MD. or D.O. degree, or in combination with other advanced degrees. The

adjusted response rate from valid returns was 63%.

The results showed that 79.3% of the faculty reported spending a half-day or less per
week on research. Even though they seem to understand the expectations to produce
research, few faculty identified a well-defined research agenda, nor currently have
multiple research projects underway. Faculty reported significantly less than one
scholarly work per year, with manuscripts for publication being the largest category of
scholarship generated. During the last two academic years, 58.2% of faculty reported
having no manuscripts accepted for publication, 67.3% no proposals/papers accepted for
conference presentations, 75.8% no national government proposals funded, and 84.3% no

national private grant proposals funded.

This study also tested a conceptual model of faculty research productivity through full
model multiple regressions for several productivity measures. Each regression model
included the following composite predictors: prior research training, psychological and
cognitive characteristics of faculty concerning research, current research environments,
resources for research, prior institutional prestige regarding research, and perceptions of
family medicine as a discipline known for research. Models also included the
demographic variables of age, years as a faculty member, hours per week spent on
research, gender, ethnicity, advanced degree, academic rank, and tenure status. Across all
forms of scholarship, the “psychological and cognitive characteristics of faculty”
composite was highly predictive of research productivity. The elements of this composite
emerge after the first faculty position and include having further developed research skills
through formalized research training, a defined research agenda, motivation to do
research, and multiple research project underway. The elements of this composite also
include having an in-depth knowledge of a research area of specialty, well-developed

professional networks, and clear expectations for promotion and tenure.

Cepyright by
JOSEPH J. BROCATO
2001

DEDICATION

To my grandparents, Raymond and Winfred Saum, who instilled in me their love for
learning, and to my parents Gene and Susan Brocato, who taught me life’s greatest lessons

and remain my guiding lights.

ACKNOWLEDGEMENTS

I would like to thank my doctoral committee, Drs. John Dirkx, James F airweather, Marilyn
Amey, and Henry Barry for their ongoing support. Special thanks is also extended to Dr.
Fairweather who spent many additional hours helping me hone this study and to Dr. Ann
Austin who provided me with my own research socialization as her graduate assistant. I
would also like to convey my sincere thanks to Drs. John Creswell from the University of
Nebraska, Carole Bland from the University of Minnesota, and Francine Hekelman and
Steven Zyzanski from Case Western Reserve University who provided invaluable mentorship
toward conceptualizing this study. I would also like to thank the American Academy of
Family Physicians Foundation for providing full funding for this research project. A big
thanks also to Dr. Brian Mavis from the Office of Medical Education Research and
Development at MSU for his statistical guidance (and friendship). Also, I greatly appreciated
the help of Dr. John Dickinson and the Association of Departments of Family Medicine who
endorsed this research study early on. I would also like to thank Mrs. Jan Szabo who helped
me mail and collate the hundreds of surveys associated with this study. Finally, I would like
to extend my heartfelt thanks to my graduate student colleagues from Michigan State
University who are not only colleagues, but lifelong friends. They include Richard Albrecht,
Kristin Anderson, Andrea Beach, Patricia Fraser, Linda Gross, Judith Kunkle, Margaret
Lafleur, Lucy Maillette, Steve Poulios, Jonathan Rohrer, Lorinda Sheppard, and Aseem

Srivastava.

vi

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................. x
LIST OF FIGURES ............................................................................................... xii
CHAPTER 1
INTRODUCTION ................................................................................................. 1
Research Activity among Family Medicine Faculty ................................ 3
The Reemergence of Primary Care Medicine and the Generalist
Physician ................................................................................................. 4
The Context of the Family Practice Specialty within Primary Care
Medicine ................................................................................................. 5
The Development of Academic Family Medicine Departments ............. 6
Challenges Toward Building Individual Faculty in Family Medicine
Departments ............................................................................................... 8
The Need for F ulltime Family Medicine Department Faculty ................ 10
Forms of Research in Family Medicine Departments .............................. 10
The Research Problem ............................................................................... 12
Study Limitations and Definitions ............................................................ 12
Study Significance ..................................................................................... 14
CHAPTER 2
REVIEW OF THE LITERATURE ....................................................................... 16
Introduction to the Review of the Literature ............................................ 16
Family Medicine Research Interests and Activities ................................. 17
Developing Family Practice Research and Family Medicine
Researchers ................................................................................................ 20
Areas of Family Practice Scholarship ............................................. 20
The Need for Family Practice Research .......................................... 22
Why Family Medicine Clinical Faculty Bear the Burden of
Scholarship ....................................................................................... 26
Overview of Faculty Research Productivity ............................................. 28
Measuring Faculty Research Productivity ...................................... 31
Correlates/Predictors of Faculty Research Productivity ................. 34
Toward a Predictive Model of Family Medicine Department Faculty
Research Productivity ................................................................................ 48
CHAPTER 3
METHODOLOGY ................................................................................................ 55
Population ................................................................................................. 57
Study Sample ............................................................................................. 59
Data Collection Strategies ......................................................................... 63
Instrumentation ................................................................................. 63

vii

Description of Study Variables ........................................................ 64

Validity Measures ............................................................................ 65
Data Collection Procedures ............................................................. 66
Data Analysis ............................................................................................. 67
Returns, Non-Retums, and Response Bias ...................................... 68
Descriptive Analysis of Independent Variables .............................. 68
Scaling of Independent and Dependent Variables .......................... 69
Statistics Used to Compare Groups, Relate Variables,
and Answer Research Questions ..................................................... 69
CHAPTER 4
RESULTS ................................................................................................. 71
Validity and Reliability Measures ............................................................. 72
Survey Response Rate Analysis ...................................................... 72
Survey Question Analysis ................................................................ 73
Descriptive Findings: Independent Variables ........................................... 75
Demographic Characteristics ........................................................... 75
Psychological and Cognitive Characteristics .................................. 80
Cumulative Advantage (Prior Research Socialization) .................. 81
Research Environments ................................................................... 82
Descriptive Findings: Family Medicine Department
Faculty Research Productivity .................................................................. 84
Two Year Mean Productivity .......................................................... 84
Productivity Differences by Demographic Variables ..................... 85
Predictors of Family Medicine Department Faculty
Research Productivity ................................................................................ 88
Results of Factor Analysis of Independent Variables ..................... 88
Results of Multiple Regression Analysis ........................................ 92
Summary ................................................................................................. 98
CHAPTER 5
DISCUSSION AND CONCLUSIONS ................................................................ 101
Introduction ................................................................................................ 101
Review of Descriptive and Predictive Findings ....................................... 101
Family Medicine Department Faculty Demographics .................... 101
Psychological and Cognitive Characteristics .................................. 103
Prior Research Socialization (Cumulative Advantage) .................. 105
Environments for Research .............................................................. 106
Family Medicine Department Faculty Research Productivity ........ 108
A Predictive Model of Family Medicine Department Faculty
Research Productivity ...................................................................... 109
Recommendations for Family Practice and Family Medicine ................. 1 15
Restructuring the Research Socialization of Future Family
Physicians and Potential Family Medicine Faculty ........................ 1 17

Drawing Strength From the Psychological and
Cognitive Characteristics of Family Medicine Faculty:

viii

The Recruitment, Selection, and Retention of Faculty ................... 120
A Reexamination of Family Medicine F aculty’s Workload

Policies (Roles and Rewards) .......................................................... 122

Revisiting This Study’s Limitations in Light of Opportunities for ..........

Future Research on Family Medicine Faculty .......................................... 123
APPENDICES ................................................................................................. 126
A. SURVEY COVER LETTER .................................................................... 127
B. SURVEY INSTRUMENT ........................................................................ 129
C. INDEPENDENT VARIABLES: FREQUENCY DISTRIBUTIONS 134
D. DEPENDENT VARIABLES: FREQUENCY DISTRIBUTIONS ......... 144
E. MEAN PRODUCTIVITY BY AGE ........................................................ 148
F. MEAN PRODUCTIVITY BY ADVANCED ACADEMIC DEGREE. 151
G. MEAN PRODUCTIVITY BY ACADEMIC RANK .............................. 153
H. MEAN PRODUCTIVITY BY YEARS OF EXPERIENCE ................... 155
I. PEARSON CORRELATIONS FOR REGRESSION MODEL

VARIABLES ............................................................................................. 157
REFERENCES ................................................................................................. 173

ix

Table
3. l

3.2

3.3

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

4.14

4.15

LIST OF TABLES

Population of US. Medical School Departments of Family Medicine
Faculty (by Advanced Academic Degree) ................................................

Family Medicine Department Faculty Study Sample Size
(by Advanced Academic Degree) .............................................................

Relationship between Research Questions, Questionnaire Items,
and Analytic Strategies ..............................................................................

Survey Response Rate (by Advanced Academic Degree Held) .............
Survey Question Wave Analysis ...............................................................

Demographic Characteristics: Age, Years of Faculty Experience, and
Hours Per Week Spent on Research Activities ........................................

Demographic Characteristics: Gender and Ethnic Origin ........................
Demographic Characteristics: Academic Rank ........................................

Demographic Characteristics: Academic Status and Non-Clinical
Advanced Degrees .....................................................................................

Academic Rank ..........................................................................................
Psychological and Cognitive Characteristics ............................................
Cumulative Advantage (Prior Research Socialization) ........................ .

Research Environments: Disciplinary, Institutional, and Departmental

Family Medicine Department Faculty Research Productivity (Means
for Last Two Academic Years) .................................................................

Factor Analysis of Independent Variables ................................................
Descriptive Statistics: Composite Variables .............................................
Summary of Full Model Multiple Regressions: Full Model Statistics

Summary of Full Model Multiple Regressions: Standardized Coefficients

Page

58

62

70
73

74

75
76

77

78
79
81
82

83

84
89
92
96

97

LIST OF TABLES (cont’d)

Table Page
5.1 Independent Variables: Frequency Distributions ..................................... 135
5.2 Dependent Variables: Frequency Distributions ........................................ 145
5.3 Mean Productivity Differences by Age .................................................... 149
5.4 Mean Productivity Differences by Advanced Academic Degree ............ 152
5.5 Mean Productivity Differences by Academic Rank ................................. 154
5.6 Mean Productivity Differences by Years of Experience .......................... 156
5.7 Pearson Correlations for Regression Model Variables ............................ 158

xi

LIST OF FIGURES

Figure Page
2.1 Conceptual Model: Predictors of Family Medicine Department
Faculty Research Productivity .................................................................. 51

xii

CHAPTER I

INTRODUCTION

Family medicine department faculty in the US. share a diversity of clinical and academic
professional roles including seeing patients as a farme physician, teaching medical
students, interns and residents as a clinical teacher, and conducting pure and applied
research contributing to the specialty and to society at-large (Bland, Simpson, Hekelman,
& Stritter, 1997). In short, family medicine department faculty are challenged to combine
all of the academic roles normally held by college faculty in mainstream higher education
while maintaining a significant patient caseload. Although much is known about family
physicians’ development as clinicians during medical school and residency (Burke,
Baron, Lemon, Losh, & Novack, 1994; Campos Outcalt & Midtling, 1993; Hosokawa &
Zweig, 1990; Merenstein & Schulte, 1990) as well as entry into new clinical positions
after residency (Reynolds, Giardino, Onady, & Siegler, 1994; Thompson et a1., 1998),
little is known about those who have decided to pursue an academic research-oriented

career track in departments of family medicine (Taylor et a1., 1991).

Developing a better understanding of research activities by faculty in family medicine is
vital for a number of reasons. From a macro societal view, clinical family medicine
faculty researchers are uniquely positioned to conduct forms of research that cannot be
filled by any other medical specialty’s faculty (Culpepper & Becker, 1987a). Because of
their positioning as the point-of first contact in patient care, as well as a focal point for

continued longitudinal care, family medicine physician faculty have a unique opportunity

to conduct pure and applied research. From a micro perspective, building a core of
research producing faculty is vital to enhance the clinical legitimacy of the relatively new
field of family practice (Doherty, Christianson, & Sussman, 1987; Pellegrino, 1987;
Stephens, 1987) as well as the academic legitimacy of family medicine (Colwill, 1987;

Culpepper & Becker, 1987a; Doherty et a1., 1987).

Geyrnan (1978b) distinguishes clinical and non-clinical (usually educational specialist)
family medicine faculty who have traditionally carried the bulk of research production in
departments of family medicine. According to Geyman, "the (clinical) family physician
must play a central role in identifying and pursuing researchable questions, drawing on
other disciplines for help as needed. Research in family practice cannot be delegated to
non-clinical researchers, and cannot be meaningful without the combined efforts of the
university and the 'real world' practice community (p.52)." Rodnick (1987) furthers this
view by identifying M.D. faculty as the key cohort toward developing “a critical mass of
(family) medicine faculty” (p.306). Rodnick states that these faculty "must overcome
their innate inhibitions and force themselves to begin academic writing at the earliest
possible moment following their decision to commit to academic careers. What they
write about and where it is published is initially of secondary importance. (p.306).”
Henry (1997) is even more specific in targeting fulltime tenure stream clinical
“researcher/teachers” in family medicine as “the group of faculty who possess a great
interest in the scholarship of discovery and are most likely to occupy tenured slots. . .”,

and accordingly , “have the greatest responsibility to pursue new knowledge (p.258)".

This study proposes to examine the research productivity of family medicine faculty in
US. medical schools. The study considers the effects of prior research experiences, the
diverse characteristics of these faculty, and the environmental conditions within medical
schools and departments of family medicine on faculty research productivity. The data
will assist family medicine leaders attempting to foster national research agendas,
departments of family medicine examining promotion and tenure guidelines, faculty
developers attempting to train family medicine faculty in research skills, and family

medicine faculty attempting to gauge baseline research productivity for the discipline.

Research Activity among Family Medicine Faculty

 

In aggregate, family medicine faculty have generally not been well-trained to conduct
research (Bland, Hitchcock, Anderson, & Stritter, 1987; Henry, 1997) or to teach medical
students (Holloway et a1., 1995). Additionally, they have not been socialized to accept
the values and attitudes of the academic profession (Bland, Schmitz, Stritter, Henry, &
Aluise, 1990; Rogers, Holloway, & Miller, 1990; Sheets & Schwenk, 1990). Instead,
family medicine faculty have conceived of themselves as practitioners first and foremost.
Gradually family medicine faculty are receiving limited academic training as more
residency programs require research projects as part of their educational process (Alguire,
Anderson, Albrecht, & Poland, 1996). Specialized faculty development fellowship
programs to train these faculty in research and teaching skills also are more evident
(Anderson, Stritter, Mygdal, Amdt, & Reid, 1997; Bland, Hitchcock, Anderson, &

Sritter, 1986; Bland et a1., 1987; Hitchcock, Anderson, Stritter, & Bland, 1988;

Hitchcock, Lamkin, Mygdal, Clarke, & Clarke, 1986; Hitchcock, Stritter, & Bland, 1993;

Hueston, 1993b; McGaghie et a1., 1990).

Yet this academic exposure is not the norm for family medicine faculty. Most of their
academic training occurs “on-the job” (Baldwin, Levine, & McCormick, 1995; Bland et
a1., 1990; Friedman, Alpert, & Green, 1994) or through in-house faculty development
programming, if at all. Although the evidence is quite limited, it appears that few family
medicine faculty engage in research more than 10% of their time, particularly those with
clinical responsibilities. Little of this is in the form of scholarly peer reviewed
publications (Hitchcock & Buck, 1990). Among the most important reasons for this

pattern is the historical development of primary care and of family practice as a specialty.

The Reemergence of Primary Care Medicine and the Generalist Physician

 

Generalist physicians are considered the line of first medical treatment in the United
States. The majority of patient visits per year occur within their offices, often prior to and
usually without subsequent referral to specialists (Geyman, 1978a). Between 90-95% of
all doctor-patient contacts occur at the primary care level (Geyman, 19783). Generalist
physicians must possess a diverse set of skills and fundamental knowledge (Cassell,
1997) that will equip them to serve a range of patients: the old and young, white and
nonwhite, those from suburban, as well as underserved and rural settings. Although some
would argue that all physicians practice varying forms of primary care (Inui et a1., 1998),

others posit that “primary care medicine can be best provided by generalists who are

specifically trained to meet the broad, as well as the intellectually and technically

exacting, demands implied in the definition.” (Cassell, 1997, p.4).

The demand for expertise in a wide range of presenting medical conditions makes it vital
that generalist physicians be properly trained within a well-defmed specialty of medicine
called primary care (Inui et a1., 1998; Schatz, Realini, & Charney, 1996). Further,

generalist physicians are increasingly being called to develop a core body of primary care

research to provide a foundation of patient-centered clinical expertise.

The Context of the Family Practice Specialty within Primary Care Medicine

 

Within primary care is a large cohesive group of generalist physicians, the family
physician practicing the newest of the 20 medical specialties called family practice.
Family physicians account for 30% of all office visits per year (Culpepper, 1993).
Additionally, each year they see more than one member of at least 90% of all families in
the US. (Fry & Gambrill, 1978). Beyond mere patient numbers, family physicians
traditionally have reached populations previously underserved, including inner city and
urban poor (Freeman, Loewe, & Benson, 1998; Xu, Veloski, Hojat, Rabinowitz, &
Rattner, 1997) as well as rural communities (Blondell, Norris, & Coombs, 1992;
Kassebaurn & Szenas, 1993). These family physicians serve as the initial point of contact
for many families, as well as serve as the major source for continuing healthcare.
Geyman (1978a) succinctly summarizes the central thrust of family practice as “the body
of knowledge and skills applied by the family physician as he/she provides primary,

continuing, and comprehensive heath care to patients and their families regardless of their

age, sex, or presenting complaint... (family practice) cuts across (the) territorial
boundaries of all traditional specialties, and varies in its application by each family
physician based upon his/her own training, interests and skills, as well as the community

in which he/she practices and the proximity to other medical resources” (p.594).

The Development of Academic Family Medicine Departments

 

As with the other medical specialties, family practice has sought to develop a core body
of teaching and research housed within academic departments in teaching hospitals and
medical schools. As of 2000, 127 departments of family medicine in the US. provide the
leadership for the clinical and academic directions of the family practice specialty.
Historically, family medicine departments have augmented a small cadre of tenure stream
fulltime faculty with non-tenure stream firll-time faculty and part-time voluntary faculty
(Bickel, 1991). However, the medical literature as well as leaders in family medicine
increasingly have called for a nucleus of full-time tenure stream family medicine faculty
as the academic foundation for the discipline (Henry, 1997). Yet several barriers confront

the development of family medicine departments as major research producing units.

One of the largest challenges facing family medicine departments is their relative
physical dispersion. Quite often family medicine faculty are scattered across several
hospitals and clinics for at least a portion of their workdays, making academic meetings
and research collaboration difficult. Centralized departmental space is becoming more
commonplace, but individual faculty often have at least two if not three or four office

sites at which they work. Coordinating faculty meetings, not to mention faculty

development programming, becomes extremely challenging for department chairs as well

as for developers seeking to reach a majority of most family medicine department faculty.

In addition, the field of family medicine has few academic foundations upon which to
build. Because of a lack of precedent for research and other forms of academic discourse,
the mandate for conducting research and improving teaching within family departments
is not universally shared by all individual faculty, including department chairs (Garr,
1986; Hueston, 1993b; Katemdahl, 1994). Nor is there universal understanding of the
value of academic development as a tool for individual faculty advancement (Applegate

& Williams, 1990; Zyzanski, Williams, Flocke, Acheson, & Kelly, 1996).

A third obstacle is that the faculty within family medicine departments differ in their
academic interests. Some faculty may be engaged in teaching and research on childhood
development while others may center their efforts on geriatric education (Culpepper &
Becker, 1987a; Parkerson & et a1., 1982). Finding ways to link faculty within and
between departments into research networks (Nutting, 1996; Woods, Reid, Amdt, Curtis,
& Stritter, 1997) and scholarly discourse networks for resource sharing remains

problematic (Mavis & Brocato, 1998).

A final hurdle is a perception within some medical specialties and subspecialties that
family medicine departments (and their faculty) possess less prestige and respect (Block,
Clark-Chiarelli, Peters, & Singer, 1996; Friedman et a1., 1994; Stephens, 1990). In

comparison to non-primary care specialty academic units, family medicine departments

have a comparatively weaker financial base (Friedman et a1., 1994). These departments
are not well represented on institutional admissions and cuniculum committees or in
faculty governance in general (Friedman et a1., 1994; Taylor et a1., 1991). Further, family
medicine faculty have a relatively poor track record in obtaining national research
funding, particularly from the National Institutes of Health (Culpepper & Becker, 1987a;
Sweeney & Jones, 1993). Small numbers of these faculty achieve tenure status (Gjerde,

1994; Jackson & MacInnes, 1984).

Challenges Toward Building Individual Faculty in Family Medicine Departments

 

In addition to challenges facing departments of family medicine, individual faculty within
these departments also confront barriers to their academic development. Family medicine
faculty on average are younger than their colleagues in other specialties of medicine
(Friedman et a1., 1994). Swee (1989) found that 75% of the family medicine faculty had
been in academia for 8 years or less, while 40% had been appointed for less than four

years. This relative inexperience has implications for research productivity.

F amily medicine faculty also are less likely than other medical faculty to have received
post-graduate teaching and research fellowships, although this gap has been narrowing
over the last decade (Bland & Stritter, 1988). Family medicine faculty often enter new
positions less well prepared for their new academic roles with the result being a reliance

on informal, “on-the-job, trial-by—error” processes.

Family medicine faculty also perceive of themselves as clinicians first and academics
second (Blackburn & Fox, 1976; Holloway, Wilkerson, & Hejduk, 1997). Medicine
faculty in general train to be medical practitioners with little if any forethought given
toward becoming future academics (Blackburn & Fox, 1976; Burke, 1992) much less to
regularly engage in research as a major role (Burke, 1992; Hueston, 1993a). This
prioritization has been reinforced by family medicine departments attempting to balance
faculty’s time to conduct scholarly activities with the need to generate patient dollars
through clinical practice (The Association of Departments of Family Medicine Task
Force on Clinical Practice in US family medicine departments in academic medical
centers,1997). The primary role perception for family medicine faculty remains one

rooted in medical practice.

Finally, few research producing academic role models exist within the field (Morzinski,
Diehr, Bower, & Simpson, 1996; Stange & Hekelman, 1990). As Bland and Schmitz
(1986) have shown, having a cohort from which to share ideas toward developing
scholarly research networks both within and outside academic units is vital toward
influencing research productivity. Because of a relative shortage of research intensive
faculty at the department level, the advancement of a research agenda in family medicine

has been painstakingly slow much to the chagrin of national family medicine leaders.

The Need for Full-Time Family Medicine Department Faculty

 

Family medicine departments need a cadre of full-time faculty pursuing both clinical and

educational research because community-based part-time family medicine faculty are not

able to develop a culture of research and scholarship for the discipline (Hueston, 1993a).
The nature of this research should be consistent with family medicine practice; therefore,
the research has the potential to be longitudinal in nature, cutting across gender, ethnic,
and familial lines toward larger populations. Thus, family medicine studies need faculty
who are willing to spend several years, even generations, within a single research project
or research focus. The continuity provided by fulltime tenure stream faculty cannot be
discounted and must be encouraged toward engaging in lines of practical clinical

research.

Forms of Research in Family Medicine Departments

 

F arnily medicine departments also must encourage research to attain credibility within
colleges and universities, not only within the medical specialties. As alluded to above, the
need for full-time tenure stream family medicine researchers is clear from the societal
benefits accrued from the research they can produce. However, the need for research in
family medicine departments also is desired because of a differing professional paradigm
embraced by family medicine researchers. According to Colwill (1986), that which
largely differentiates family medicine researchers has both conceptual and philosophical
roots operationalized in the approach taken toward conducting research. More
specifically to Colwill, family medicine researchers focus on the bond between patient
and doctor, have generational attentions centered within the family unit, emphasize
educational principles as guides, and are mindful of the patient as both an individual as

well as part of families and communities.

10

Phillips (1978) posits further that family medicine research includes a unique melding of
disciplines (such as biomedical sciences, epidemiology, behavioral science, philosophy,
and ethics), research foci (such as pathogenesis and mechanisms, clinical strategies,
health services, and public policy), and a diverse mix of research methods (such as multi-
centered studies, collaborative models, as well as solo research) unfulfilled in other
branches of medicine. Geyman (1978b) points out that family medicine research
addresses topical, relevant, and applied issues for healthcare including a focus on
managed care, preventive medicine, the effects of diagnostic and therapeutic methods,

the long-term outcomes of healthcare, as well as educational issues.

In sum, Culpepper (1991) provides a snapshot of the unique niche filled by this research
when he states that “family medicine research addresses the need for knowledge by
family physicians so they may better manage their patients, their families, and their
practices and fulfill their healthcare role at the community level. Further, family medicine
research particularly seeks to answer the questions which require the family practice
setting or the relationship among family physician, patient, farme and community. It

investigates issues from the family physician and patient perspectives.” (p.10)

The Research Problem

 

As part of this study, five central questions emerged as guideposts. They include:
(1) What are the individual faculty characteristics (demographic, psychological &

cognitive) of family medicine department faculty?

11

(2) What prior socializing experiences in terms of research have family medicine

department faculty had before accepting their first faculty position?

(3) What are the characteristics of family medicine department faculty's research

environments?

(4) What is the research productivity of family medicine department faculty in US.

medical schools?
(5) What prior socializing experiences, individual faculty characteristics, and
characteristics of research environments of family medicine faculty predict research

productivity?

Study Limitations and Definitions

 

This study focused on US. medical school family medicine department full-time faculty
who hold the MD or DC as the only advanced degree, or hold the MD. or DC. in
combination with another advanced degree. As alluded to earlier, the reason for this
limitation is because, as specified by both Rodnick (1987) and Henry (1997), these
family medicine faculty provide the foundations of a clinical research core for the
discipline. Although Ph.D.-degreed faculty produce meaningful educational research for
departments of family medicine, they are the smallest group of faculty by academic
degree comprising only 13% of all family medicine department faculty (The Association

of American Medical Colleges, 1999). Additionally, Ph.D.-degreed family medicine

12

faculty possess an entirely different socialization than do clinical faculty, including more
formal education in research skills as well as experience conducting research (Aren &
Ben-David, 1968; Bland et a1., 1990). Further, the resources and time available to
conduct research for this group within departments of family medicine may be vastly

different from clinical family medicine faculty.

This study also was limited to US. medical school’s departments of family medicine
faculty rather than including additional family medicine faculty at community-based
hospitals or clinics. These community-based family medicine faculty normally carry
significantly higher clinical and teaching loads and much smaller research responsibilities
and expectations (Garr, 1986; Hitchcock & Buck, 1990), making it problematic to

include them in this study.

For purposes of this study, I defined research productivity as the number of peer-
reviewed journal articles, national conference presentations, and national grants (both
private and public) over the most recent two-year period. The use of a two year period of
self-reporting was selected as suggested by Creswell (1985a) as a suitable measure of
faculty productivity studies. A “peer reviewed” focus was selected as a proxy for quality.
Research productivity areas are limited to the three highly traditional forms of peer-
reviewed scholarship because of the need for family medicine faculty to compile a core
body of research comparable to other medical specialties. These traditional forms of
research productivity are aligned to the three-step process proposed for developing a core

body of research in family medicine by Dehaven, et.al. (1994) that includes "(grant)

13

funding, presenting, and publishing (p.307)". Additionally, according to Applegate
(1990) these three forms of research productivity are vitally important for the career
advancement of academic medicine faculty, particularly those seeking promotion and

tenure.

Study Significance

 

This study is highly useful for multiple audiences in medical education through a
aggregate measurement of the research productivity of family medicine faculty as well as
the factors that may influence their research productivity. The study is also useful for
national family medicine leaders seeking to encourage research within the specialty of
family practice to move the field toward a more scholarly definition and standing, as well

as to meet the needs nationally for community-based research with clear societal benefits.

Additionally, the information garnered in this study is informative for deans of medical
schools seeking to review promotion and tenure guidelines for their schools and colleges.
Deciding what role research can and should play in the process can be better informed by
understanding how these faculty have been trained to do research before and after
becoming faculty members as well as the resources they have at their disposal to conduct

research.

Further, this study provides family medicine department chairs with a snapshot of the

psychological and cognitive characteristics of family medicine faculty relating to research

as well as a measurement of the research environments for these faculty. This information

14

is useful to department chairs toward analyzing the scholarly expectations for their

departments as well as obtaining quantifiable data for promotion and tenure guidelines.

Finally, individual faculty within family medicine departments are provided with a
composite of the conditions that may influence their own potential research success, and
ultimately, promotion and tenure where applicable. By having a better understanding of
both baseline levels of performance, as well as the “gold standard” for research
productivity of faculty nationally, family medicine faculty are equipped with both a
guidepost for their longitudinal academic development as well as a measuring stick for

their academic careers.

15

CHAPTER 11

REVIEW OF THE LITERATURE

Introduction
Although no studies have examined the research productivity of family medicine
department faculty, the literature does enable developing elements useful toward
conceptualizing a predictive framework of family medicine department faculty’s research
productivity. The review of the literature highlights these foundational studies and is
presented in the following four sections: (1) past studies of family medicine faculty’s
research interests and activities, (2) the medical specialty of family practice and the
academic discipline of family medicine’s efforts toward developing faculty research, (3)
faculty research productivity, and (4) the correlates of faculty research productivity. The
review of the literature ends with a conceptual framework of family medicine department

faculty’s research productivity that serves as a guide for this study.

16

Family Medicine Research Interests and Activities
Although much has been written in the family medicine literature concerning the need for
family medicine faculty to produce research, only a few empirical studies examine family
medicine faculty’s research interests and activities. One of the first major studies of
family medicine faculty was conducted by Gjerde, Clements, and Clements (1982). They
identified 74 family medicine department faculty (holding either clinical and/or
academic degrees) in 104 departments nationally nominated for promotion during 1980-
81 to determine their publication characteriStics. The results of the departmental survey
showed a relatively low level of research productivity for family medicine faculty
seeking promotion. On average, promoted assistant professor candidates in the study
reported 0.6 journal articles published prior to promotion, 7.4 journal articles published
for associate professors, and 11.0 journal articles published for full professors—well less
than one journal article per academic year. Additionally, 28% of nominees reported not
being published in any journals at all. Despite relatively low levels of research
productivity, 65% of the nominated faculty achieved promotion that academic year.
Gjerde’s (1994) follow-up during the 1988-89 academic year found similar results with
one exception-- productivity at the promoted assistant professor level was higher (2.7
articles). Non-clinical faculty candidates reported significantly more publications than did
clinical faculty. Successful candidates for promotion published more articles than did

non-successful candidates.

17

Culpepper and Franks (1983) conducted the first national survey of family medicine
research activities in departments of family medicine and family medicine residencies.
Their survey of 353 family medicine departments and residencies found that 19% of
departments and residencies had little or no emphasis on research, 44% reported an
“emerging emphasis on research”, while 37% had a “visible emphasis on research.” The
number of physician faculty involved in research also was quite low as reported by
departments and residencies: 66% of faculty reported as having less then 10% of their
time devoted to research, 26% spending between 10-50%, and 7.8% dedicating greater
than 50% of their time on research. These findings were consistent with a similar 1982
study conducted by The Study Group on Family Medicine Research (Parkerson & et a1.,

1982).

The Culpepper and Franks study also described several impediments to research reported
by departments or units. These included “lack of faculty time”, “lack of funding for
faculty time for research”, “lack of funding for staff, equipment, and supplies, and “lack

of faculty research skills”.

In 1986, Garr (1986) surveyed 986 family physicians teaching fulltime in 214 family
medicine residency programs to determine relationships between demographic
characteristics and job satisfaction. Only 22% were satisfied with their opportunities to
do research. Additionally, only 18% were satisfied with their own research skills. Also,
only 18% of the faculty stated they were satisfied with the quality of their research

efforts.

18

Nearly a decade after the Culpepper and Frank study, Hueston (1993b) surveyed 208
fulltime physician faculty in family practice residency programs focusing on their
demographics, training and experiences, and factors that motivated them to select
academic careers. Hueston found that over 80% of family medicine faculty spent less
than 10% of their time on research. About one-third spend no time at all on scholarship.
Additionally, those that did devote more than 10% of their time doing research were
more likely to have been fellowship trained, employed in university training programs,
have more academic experience, and identify opportunities to do research. Hueston found
that having an interest in research when first seeking a faculty position was the greatest

influence toward devoting more time to research for family medicine faculty.

The most recent published study of family medicine faculty research activity was by
Zyzanski, et. a1. (1996) who examined the publication patterns of successful candidates
for promotion and tenure to associate professor in family medicine departments
nationally. Zyzanski, et.al found that successful associate professor candidates averaged
7.6 peer-reviewed journal publications (or 1.3 per year), 44% had at least one funded

research grant, and 54% had at least one funded training grant.

The results show that a large majority of family medicine faculty are producing little if

any research despite increased emphasis on research by leaders in family medicine. The

next section of this review synthesizes the literature from family medicine and family

19

practice concerning the need for family physicians to conduct scholarship, as well as the

various research roles available for family physicians.

Developing Family Practice Research and Family Medicine Researchers
The literature on family practice focuses on kinds of research family physicians could
produce. Additionally, it describes why these bodies of research are vital toward
contributing to the development of the specialty of family practice and the legitimacy of
the field of study. The sections that follow summarize the need for family physicians to
engage in research and how academic family medicine departments and their faculty

shoulder the burden for this scholarship.

Areas of Family Practice Scholarship

According to Geyman (1978a), family practice research covers three areas: clinical
strategies, health care services, and educational methods. Specific examples include cost
effectiveness of health maintenance and prevention procedures, functional outcomes of

care, and effectiveness of educational approaches at various learning levels.

According to Stephens (1982), the value of family practice research is in bridging the gap
between primary care and medical subspecialties. These subspecialties of shared research
responsibility include child care and growth and development (shared with pediatrics),
early diagnosis and treatment of cancer (shared with oncology), informed patient decision
making regarding surgery (shared with general surgery), and diagnosis of mental illness

(with psychiatry).

20

Hankey (1987) developed five general areas of family practice research and estimated the
percentage of time the specialty devotes to these areas in the literature. These are “who
are we and what do we do?” (20%), “how do we teach what we do?” (21%), “how can we
survive as a specialty?” (11%), various psychosocial and biopsychosocial behaviors
(19%), and biomedical diseases and their treatment (30%). Hankey distinguished the
subspecialties from family practice by the subspecialties’ ability to produce biomedical

research well beyond the 30% produced by family practice researchers.

Building upon Stephen’s conceptualization of family practice research as boundary-
spanning, Culpepper and Becker (1987b) defined family practice research as either of
practical applicability and/or biopsychosocial integration. Practical applicability studies
often “involve illnesses commonly encountered by family physicians, and problems
which reflect the unique features of primary care practice.” These include “the
presentation of illness at an earlier stage. . .the focus on prevention as well as diagnosis
and treatment. . .the longitudinal care of patients. . .and the integration of care for a
number of diverse problems.” (p. 142). According to Culpepper and Becker, the
biopsychosocial integration theme of research moves beyond questions adhering to a
“strict biomedical paradigm” to include an integration of “individual perceptions, feelings
and values, and with interactions between the individual and his or her larger context.”

(p.142).

21

The choice of research topics is strongly influenced by the unique niche family
physicians fill. This niche includes providing healthcare that is longitudinal rather
episodic in nature, generation-spanning from infant care to geriatric care, and is also
provided to under-served and rural populations. Family physicians “point of first (and
usually last) contact” with patients, ability to provide long-term continuity of care, and
emphasis on universal healthcare, opens up practical research possibilities available to no
other non-primary care medical specialty’s physician group (Culpepper, 1993; Geyman,

1978b)

The Need for Family Practice Research

Three primary research themes are readily identifiable from the family practice and
family medicine literature. These include clarifying the identity of family practice among
other medical specialties, providing a concrete core curricula rooted in primary care
research to form the academic discipline of family medicine, and answering society’s
needs for primary care research. Leaders in family medicine view defining both the
academic discipline of family medicine and the specialty of family practice through
research as an interwoven process. As Geyman (1971) states “though much of family
medicine is derived fi'om other disciplines, all of family medicine is unique in terms of its
multi-disciplined approach to the care of the individual patient and his family. . .we
should be aware that the manner in which we define family medicine as an academic

discipline relates directly to the nature of the future field of family practice.” (p.820).

22

As evidence of this interwoven relationship between the specialty of farrrily practice and
the academic discipline of family medicine, Geyman (1977) notes two distinct phases of
family practice development. The first occurred in the 1960’s through the development of
the specialty of family practice and the emergence of academic departments of family
medicine primarily to teach in the new specialty at US. medical schools. Relevant
influences include the 1967 Millis Commission identification of “primary physician” as a
specialty-trained role, the 1969 approval of family practice as a board specialty, and the
development of residency programs and departments of family medicine in the late
1960’s and early 1970’s (Holloway et a1., 1997). Stephens (1979) firrther described the
first phase as “countercultur ” arising out society’s need for broad access to healthcare
centered in the humanistic qualities of the physician. According to Scherger (1997),
during this time “newly formed department of family medicine derived their value and
presence by having different priorities than other departments, including humanistic

education, social research, and community-based clinics.” (p.439).

The second phase of family practice development occurred at the end of the 1970’s.
According to Geyman (1977), this period “involv(ed) the better definition of the
academic discipline (of family medicine) and the development of the research base in the
field. . .therein lie(s) the life blood required to assure the continued development of family
practice as a specialty”. (p.15) In 1978, section 780 of Public Health Service Act
provided much needed government funding to improve the research skills of family
medicine faculty. These training programs allowed academic family medicine and the

specialty of family practice to begin what Scherger (1997) called the “parity phase”. In

23

addition, the Robert Wood Johnson fellowship program encouraged teaching research
skills to family practice residents. This phase sought to place family medicine on equal
footing with other academic departments in US. medical schools, and family practice to
balance between its counter-cultural roots and its need for a traditional base of

scholarship.

Scherger (1997) declared that academic family medicine is now entering the “integration
phase”. This phase involves the coordination of research and clinical practice between
family medicine and the other medical specialties, especially the other primary care
specialties of general internal medicine, general pediatrics, and general obstetrics and
gynecology. The aim of such coordination is “primary care research networks” linking

researchers between and among the medical specialties.

According to the Research Subcommittee of the Academic Family Medicine
Organizations Steering Committee, (a major conglomeration of the major academic
family medicine organizations), there are four core values for family medicine research
including three involving a societal mandates (Rogers, 1995) (p.180). According to the
Research Subcommittee, family medicine research should “advocate for an improvement
in the health of individuals, families, and communities.” Further, through family
medicine research the “incorporation of new knowledge central to quality practice can
improve decision making by patients, caregivers, learners, and policy makers”. Lastly,

family physicians conducting research allows for an “accepting (of) our leadership

24

responsibility in asking and answering important questions, because of our unique

position as integrators of healthcare”.

According to Sweeney and Jones (1993) the need for primary care research is vital to
counterbalance the predominate focus of the National Institutes of Health (NIH) on
biomedical research. Sweeney and Jones provide a compelling account of the historical
imbalance between biomedical research, dominated by the non-primary care specialties,
and mainstream family practice research (p.37):
For the past 30 years, over 95% of all medical conditions have been evaluated and
treated outside of hospitals. However, the traditional focus of medical education
and research has been on medical problems in referred and hospitalized patients.
Thus, the training of physicians and the research agenda have focused almost

exclusively on inpatient rather than outpatient evaluation and treatment.

The undifferentiated problems of patients seeing generalist physicians are met too
often with a knowledge base derived from referred or hospitalized patients, which
may not be relevant to the entry level of medical service and treatment. Office-
based, community-oriented family practice and primary care research on the
interaction between patients and physicians will enhance the ability of physicians to

more effectively care for patients.

Finally, at the Keystone Family Medicine Conference, Stange, Miller and McWhinney

(2000) listed five contributions of family medicine research:

25

1. To contribute to the growth and development of the academic discipline of family
medicine

2. To teach new generations of family practitioners “tacit knowledge”

3. To improve patient care by embracing the life course of individuals, families and
communities

4. To add to the knowledge base of the specialty of family practice

5. To address new challenges in the health care system, and in society at-large, through

the use of a biopsychosocial model of research.

Why Family Medicine Clinical Faculty Bear the Burden of Scholarship

The literature in family medicine clearly isolates the need for both tenure-stream and
non-tenure stream clinical faculty (i.e., those that hold the MD. or DC. degree rather
than the Ph.D.) to take a leadership role in the generation of scholarship. According to
Hitchcock and Buck (1990) tenure-stream family medicine faculty members are more
likely to have protected time for research and may have greater access to grant funding
opportunities. Further, tenure-track faculty should have at their disposal ancillary
research support staff resources such as secretaries, librarians, research assistants, and
statisticians to assist in research efforts. Finally, tenure-track faculty serve as mentors for
medical students, interns, residents, and junior faculty. According to Henry (1997) this
tenure-stream group can be labeled “researcher/teachers” and should “possess a great

interest in the scholarship of discovery” (p.258). The numbers of tenure-system family

26

medicine department faculty identified by Henry, however, is exceedingly small. The

aggregate volume of research they can produce is linrited by a lack of “critical mass”.

As one consequence, community-based faculty must produce scholarship about family
medicine. Although this group has not historically been charged with these
responsibilities, Hitchcock and Buck point out that they are increasingly being called to
produce research because they comprise the majority of family medicine faculty and
because of the addition of research requirements within community-based residency
programs. Hitchcock and Buck still believe that the primary research roles for this group
are to be informed consumers of research and to collaborate with medical school-based

tenure system faculty in research projects.

In contrast, Geyman (1978b) argues that tenure status is not all-important when
determining research responsibility. Instead, clinical family medicine faculty (i.e., those
that hold the D0. of MD. degree) should take leadership for research because of their
ability to conceptualize patient-centered, practice-based research questions. Further,
Geyman views clinical family medicine faculty as bridging the gap between academic
family medicine and the “real world practice community”. In sum, both tenure and non-
tenure system clinical faculty in family medicine departments shoulder the responsibility
for research because of their unique positioning within academe and clinical practice

because of their access to valuable resources for research.

27

Overview of Faculty Research Productivity
In her review of the literature on faculty research productivity Fox (1983) found one
generalization across disciplines: the average rate of faculty publication tends to be low
and the variation in performance levels remains very high. Historically, 10% of faculty
have been thought to produce 90% of all publications (Finkelstein, 1984; Ladd, 1979).
Stated further by Creswell (1985a), “why some faculty produce research year after year

and others do not is a “puzzle” (p.iii).”

When considering publication productivity of faculty, Braxton and Bayer (1986)
distinguish research and scholarship. Scholarship is both a process and a product,
involving the application of professional knowledge and skill as its primary attribute.
Although it can involve traditional research, scholarship includes broader forms of
analysis and synthesis (Boyer, 1990), Braskarnp (1994), and Glasick, Huber, and Maeroff

(1997).

According to Braxton and Bayer, traditional research involves a more narrow, centralized
focus than scholarship. Research includes the attributes of disciplined inquiry in terms of
the process, and original discovery as an end product. Bowen and Schuster (1986) believe
that this disciplined, scientific research can be in two forms: basic research that seeks to
“discover the laws of nature regardless of practical applicability”, and applied research
that involves “way of putting (basic knowledge) to practical use (p.16).” Further, they
point out that higher education institutions and college faculty, especially in “pure fields”

such as the sciences, historically have emphasized basic over applied research.

28

Logan Wilson (1942) conducted the first substantial faculty work study in the early
1940’s. In The Academic Man, Wilson examined various aspects of the work lives of
academics, including their research and teaching loads. Concerning research
performance, Wilson concluded that faculty devoting a larger percentage of time strictly
to teaching were promoted more quickly than those devoted a larger concentration of
time on research. In fact, those who devoted more time to teaching also became more
valued institutionally, as evidenced by overall performance appraisals (Creswell, 1985a).
Even decades ago, scholarly productivity could be considered a system of tradeoffs.
Faculty work allocation was, and still is, influenced by the cultures of the discipline,

institution, and department (Austin, 1990a).

According to Geiger, the development of the modern research university in the early
1900’s in America served as focal point for the generation of knowledge and a cadre of
research producing academics (Geiger, 1986). While attending to the education of
undergraduates, these research universities differentiated themselves from mainstream
higher education institutions by their additional emphasis on graduate education, ability
to recruit, retain, and further develop research producing faculty, as well as proficiency at

obtaining “extramural funding” for research activities through federal and foundation

grants.

Much has happened in academe to shape faculty research productivity during and after

World War H. Milestones include the scientific revolution precipitated by the launch of

29

Sputnik, the emergence and rapid growth of populace postsecondary education as
mandated by the GI Bill, and attention to research focusing on social and global issues
borne from the unrest of the 1960’s (Geiger, 1993; Jencks, 1977). These forces among
others have shifted how colleges and universities and their faculty conceptualize and
engage in their research activities in terms of foci for research, how research is frmded, as

well as patterns of scholarly dissemination.

Merton (1968, 1973;1957) showed that the sciences led the move toward an emphasis on
disciplinary research. Merton “studied the norms associated with scientific work in
science, patterns of competition among scientists, the reward structure of science,
scholarly refereeing, and inequality in scientific performance (In Creswell, 1985a) (p.3)”.
One of Merton’s greatest contributions to the study of research productivity was his
seminal work on the “Matthew Effect” in the sciences. According to Merton, the
Matthew Effect (1973) “consists of the accruing of greater increments of recognition for
particular scientific contributions to scientists of considerable repute, and the withholding
of such recognition fi'om scientists who have not made their mark.” (p.446). What Merton
describes is a classification of researchers into two groups, those that “have” and those
that “have not”. Those that “have” can be characterized as possessing an accumulation of
prior research socialization that includes an understanding of the expectations for
research of a particular discipline, as well as having well-developed research project
experience. Further, the advantaged “haves” have had considerable reward and

recognition for their research that is extrinsically motivating in nature. This prior research

30

socialization is what Creswell (1985b) and Fox (1983) described later as “cumulative

advantage”.

Measuring Faculty Research Productivity

Several uniform measures of faculty research productivity have been mentioned in the
higher and medical education literature. In his synthesis of the faculty research
productivity literature from nearly every discipline in higher education, Creswell (1985a)
found that the three most frequent measures of faculty research performance were
publication counts, citation counts, and peer and colleague ratings. According to
Creswell, each of these performance measures addresses a different dimension of faculty
research productivity. Counts of publications measure the quantity of an individual’s
research output, citations assess the quality of the publications, and peer or colleague

ratings measure the value of the contributions of research to the discipline.

Faculty publication counts can either be “straight counts” or “weighted counts” (Collins,
1993). Perhaps the easiest way to gather counts is to ask respondents to self-report the
number of publications produced for a particular period of time. Counting all publications
equally may be simplistic because it ignores the quality of the publications. For example,
a co-authored, non-peer reviewed essay could be counted as equivalent to a peer-
reviewed, first-authored journal article. Further problems could arise when equal weight
is given to many of the peer-reviewed publications in newer journals whose review
standards may be less rigorous than more established journals. In sum, the main

advantage to the researcher in employing self-reported straight counts of publications lies

31

in the ease of data collection. Further, publication counts are often useful for making
comparisons across various disciplines. Finally, self reported publication count data are
highly reliable when compared to verified publication counts (Creamer, 1998; Creswell,

1985b)

One method of adding quality into self-reported counts is to define eligible publications
carefully. Faculty members can be asked to list non-refereed publications separately from
referred journals. Single authored papers can be distinguished from multiple-authored
ones. The types of publications—joumal articles, books, monographs, book reviews—
also can be distinguished. In this manner, quality can be taken into account by including

only publications meeting higher prestige criteria.

Assigning weights to publications based upon predetermined criteria is another way to
take quality into account. Weights can be based upon the reputation of a particular
journal, by authorship position, type of publication, peer review status, or by
combinations of these approaches (Hilton, Fisher, Lopez, & Sanders, 1997; Print &
Hattie, 1997; Schneid, Hamm, & Crawford, 2000). Hypothetically, a weighted approach
provides a more accurate measure of research productivity. In practice, it is cumbersome
to apply. Weighting schemes also are seldom comparable across disciplines. Further,
reaching a consensus on weight values remains highly contentious by all counts

(Creswell, 1985a).

32

Citation counts also have been used to measure faculty research productivity, although
collectively this data is cumbersome (Braskamp & Ory, 1994; Creamer, 1998; Creswell,
1985b). One way of gathering citation data is by obtaining curriculum vitae from faculty
and verifying listed citations through citation abstracts and databases. More recently,
streamlined discipline-specific indices have been developed to provide research
summaries along with a cross reference of all authors cited in these works. According to
Centra (1981), citation data better reflects the impact of faculty work. Unfortunately,
citation counts tend to encourage the overuse of self-citations and citations of fiiend’s
works. Further, citation counts do not distinguish between positive and negative
comments about the work being cited. Also, citation indices are subject to a long lag-time

because of the long peer review and publication process.

In addition to publications, other measures of traditional forms of scholarship include
grants, conference presentations, and professional reports (Centra, 1981; Creswell,
1985b). Boyer’s (1990) seminal work called “Scholarship Reconsidered: Priorities of the
Professoriate” has encouraged reconceptualizing faculty research into a broader vision of
a four types of scholarship: the scholarships of discovery, integration, application, and

teaching.

Boyer’s broader conceptualization has not yet displaced the traditional peer-reviewed
publication model of faculty productivity in promotion and tenure decisions within
academe. Additionally, broader views of scholarship are not easily comparable within

and across disciplinary boundaries toward defining a core body of knowledge. Thus,

33

faculty research productivity measured broadly may not be as useful as traditional
scholarship measures--particularly with underdeveloped academic disciplines seeking to

establish legitimacy.

Faculty research productivity studies in medicine and health also have used a variety of
traditional research productivity measures with self-reported publication counts being the
most commonplace (Collins, 1993). Hillman and Fajardo (1989) conducted a national
survey of academic radiologists to examine research productivity by straight counts of
peer reviewed publications. Vardan and Smulyan (1990) did the same with faculty in
general internal medicine. In contrast, Ellwein and Khachab (1989) used weighted counts
to assess research productivity for both clinical and basic science departments within one
institution. Finally, Hekelman, Zyzanski, and F locke (1995) used a straight count of peer
reviewed publications and grants over a two year period to assess the research

productivity of new faculty at Case Western Reserve University.

Correlates/Predictors of Faculty Research Productivity in Higher and Medical
Education

One way of classifying the correlates of faculty research productivity is to divide them
into separate groups based on the same theoretical constructs(Collins, 1993). In her study
of science faculty, Fox (1983) includes three categories of correlates of research

productivity: individual level characteristics (such as psychological characteristics, work

34

habits, and demographics), work environmental factors, and reinforcing feedback.

Creswell (1985a, 1985b, 1990a, 1990b) expanded this list as follows (1985, p.15-24):

(1) Psychological and individual factors: productive researchers possess certain
psychological and individual characteristics that exceed less productive researchers
includes higher intellectual ability, more motivation to perform, a certain type of
personality or cognitive structure, and particular background characteristics such as
age and gender.

(2) Cumulative advantage: productive researchers have amassed resources and
positioning in their preparatory training and early in their career which predisposes
them to be more successful such as training in a research intensive department or
institution which has major sources of resource funding.

(3) Reinforcement: productive researchers are provided with frequent and consistent
praise for their research both from colleagues and from mentors throughout one’s
development as a researcher.

(4) Disciplinary norms: productive researchers are those who reflect the research
priorities of their specialty areas through their production level, mix of research
topics, as well as foci for research. Additional, they reflect the methodological norms

that accepted with disciplines.

Faculty research productivity studies rarely use these categories of factors together in a

predictive, conceptual manner due to the breadth and depth of the factors involved

(Collins, 1993; Jungrrickel & Creswell, 1994). However, notable exceptions include

35

Blackburn and Lawrence (1995) and Fairweather and Rhoads (1995) studies of higher
education faculty research productivity. Without such a broad framework, determining
the relative value of one category of factors over another is problematic. The following
sections synthesize the faculty research productivity literature using Creswell’s categories

as a guide.

Psychological and individual factors

Psychological and individual factors used in studies of faculty research productivity
include intelligence, motivation, personality characteristics, stress, age, gender, ethnicity,
advanced degree, acaderrric rank, and tenure status (Creswell, 1985b). Intrinsic
motivation is highly correlated with faculty research productivity. Science faculty often
are motivated to pursue their own lines of research (i.e., forming a research niche) in part
by the satisfaction of producing new discoveries (Pelz & Andrews, 1966). This drive to
conduct research, called the "sacred spar ” (Merton, 1973), is seen as pushing high
producing faculty toward further discovery. Behmeyer (1974), found that these intrinsic
factors were the strongest predictors of faculty research productivity, a pattern also found
by Harrington and Levine with dental school faculty (1986) and Rebne (1989) with

faculty across multiple disciplines.

The role of stress in research productivity is more ambiguous (Behmeyer, 1974;
Behmeyer & Blackburn, 1975; Blackburn & Bentley, 1993; Blau, 1973). Pelz and
Andrews (1966) described stress as resulting from the difference between preferred time

spent on research and actual time spent on research. Although faculty workload studies

36

Show a significant difference between faculty’s preferred effort for research and the
actual time they spend on research (Blackburn & Lawrence, 1995; Finkelstein, 1984;
Finkelstein, Seal, & Schuster, 1998), the relationship between this discrepancy and

research productivity is minimal.

Most faculty productivity studies have shown that the relationship between career
publications and age is not linear, although the overall rate of publication in general
declines with age (Behmeyer, 1974; Behmeyer & Blackburn, 1975; Finkelstein, 1984;
Finkelstein et a1., 1998; Lawrence & Blackburn, 1988). However, an examination of
national faculty research productivity by F airweather (1996) did show such a linear
relationship between career publications and age to be present. Yet, most studies isolate
age as being a mediating influence on other predictors of research productivity, such as in
time spent on research and motivation to conduct research, rather than being directly

predictive (Collins, 1993; Creswell, 1985b).

As for gender, men historically have had higher levels of research productivity than
women (Blackburn & Lawrence, 1995; Creamer, 1998; Finkelstein, 1984; Finkelstein et
a1., 1998). This gender gap has narrowed significantly in recent years (Blackburn &
Lawrence, 1995). Similarly, non-minority faculty produce more research than minority
faculty. The literature suggests that this gap is not narrowing over time (Creamer, 1998).
Most studies of the relationship between gender and ethnicity with faculty research

productivity focus on the career rather than on shorter time periods (Bentley &

37

Blackburn, 1990; Blackburn, Wenzel, & Bieber, 1994; Cole & Singer, 1991; Creamer,

1998; Long, 1990; Nettles & Pema, 1995).

Some authors believe that white males have better established research networks within
and outside their departments than do women and minorities (Finkelstein, 1984;
Hitchcock, Bland, Hekelman, & Blumenthal, 1995). Further, women and minority faculty
are over-represented in the “soft science areas” with traditionally lower expectations for
research (Finkelstein, 1984). Women and minority faculty are less likely to be employed
at research-oriented universities and spend more of their time teaching-often at the
expense of research activities (Bowen & Schuster, 1986). Finally, women faculty often
have family demands that can detract from time available for research (Creamer, 1998).
As with age, gender and ethnicity are considered mediating influences on productivity
rather than direct causes (Blackburn et a1., 1994; Cole & Singer, 1991; Creamer, 1998):
“race and gender do not have a direct effect on publication productivity, but an indirect
effect through factors such as rank and academic field; institutional factors, including
work assignment; and environmental factors, such as access to funding and influential

collegial networks (Creamer, 1998, p.3).”

Advanced degree, academic rank, and tenure status are moderately, positively related to
publication output. In the health professions, such as nursing and dentistry, possessing a
Ph.D. degree has been associated with higher levels of faculty reSearch productivity in

most studies (Collins, 1993; Flanigan et a1., 1988; Harrington & Levine, 1986; Kraemer

& Lyons, 1989; Megel, 1985). Acquiring the Ph.D. apparently teaches health professions

38

faculty the academic norms and values as well as research skills that graduate students

acquire in other graduate programs.

In most studies of faculty research productivity, academic rank and tenure status are
positively indicative of scholarly output (Blackburn & Lawrence, 1995). This pattern is
less obvious in the health professions where faculty are less likely to hold tenure status.
Finkelstein (1984) believes that academic rank is a significant predictor of the rate of
publication because faculty in higher ranks have more control over their workload
assignment. Additionally, Finkelstein as well as Fulton and Trow (1974) believe that
senior ranks contain faculty who succeed at research; those who remain are more
productive as a whole. Tenure status is not related to faculty research productivity
(Blackburn, Behmeyer, & Hall, 1978; Megel, 1985; Neumann, 1979). Reskin (1977) says
that “while there are powerful sanctions in the long run, promotion, tenure, and salary
raises are probably not especially effective in maintaining day-to-day conformity to

productivity norms (p.491)”.

Of the various psychological and cognitive factors, including intelligence, motivation,
personality characteristics, stress, age, gender, only intrinsic motivation has consistently
been identified as a strong predictor of faculty research productivity. Most of the other
psychological and cognitive factors have shown a modest correlation with faculty
research productivity. They are better seen as mediating effects. The use of broad
psychological/cognitive factors, other than intrinsic motivation, seems of limited value in

developing a discipline-specific, predictive model of faculty research productivity.

39

Cumulative advantage (Research Socialization)

According to Merton (1973), faculty in most higher education disciplines acquire
research recognition and resources through their prior academic and professional training.
These experiences make it easier for them to acquire additional recognition and resources
in the future. This cumulative advantage, called anticipatory socialization, provides the
elements of the first stage of a faculty member’s socialization to academe (Tierney &
Rhoads, 1993; Van Maanen, 1976). The second phase, organizational socialization,
occurs from faculty recruitment through accepting the first faculty position and beginning

faculty employment.

During graduate school future faculty are socialized to a research culture. Anticipatory
socialization is as an important concept in describing the development of graduate
students as future academics. Gotlieb (1968) described how departmental climates and
cultures, as well as interactions with faculty, influence future career choices of graduate
students including their research foci. Bess (1978) described how certain graduate
students may be predisposed toward faculty careers by developing a set of distinctive
academic values, attitudes and beliefs. Weidman and Stein’s (1990) conceptual model for
graduate students' professional socialization included socializing agents and events
including student's background characteristics, educational experiences, non-educational
reference groups (family, employer, peer group), professional associations and
professional practice. The outcomes of this graduate student socialization include

developed and shared values and attitudes, academic aspirations, academic norms,

40

conceptualization of professional authority and status, development of knowledge and

skills, individual identity, and professional commitment.

Blackburn and Fox (1976) conducted the only major study of how medical faculty are
academically socialized through patterns of accumulative advantage. Blackburn and F ox
found that very few clinical medical faculty (i.e. medical faculty with the MD. degree)
aspire to academic careers during their prior academic and professional training. Further,
these clinical medical faculty made decisions about choosing an academic career
relatively late in comparison to non-clinical medical faculty (i.e. medical faculty with the
Ph.D. degree). Blackburn and Fox also showed that non-clinical medical faculty are
more likely than clinical faculty to accept "university professorial norms". Clinical
faculty are more accepting of new clinical recruits to the professorate. Regardless of
terminal degree, however, Blackburn and Fox found that the medical faculty in their
study exhibited similar academic goals, suggesting that their socialization to academic

roles occurs on the job rather than during acaderrric training.

Whatever the motivation, once faculty members publish thel develop a cumulative
advantage, highly predictive of future research productivity (Allison & Stewart, 1974;
Blackburn et a1., 1978; Clemente, 1973; Collins, 1993; Creswell, 1985b; Fox, 1983;
Megel, 1985). This accumulative advantage includes early success in publishing, prior
research project experience, research mentorship, development of research skills,

collaboration on research projects, and research sponsorship (Collins, 1993; Creswell,

41

1985b; Fox, 1983, 1996). In sum, cumulative advantage is highly predictive of faculty

research productivity.

Environmental Socialization and Reinforcement: The Influence of Disciplinary,
Institutional and Departmental Cultural Norms for Research

Academic environments and their attendant cultures or climates (Peterson & Spencer,
1990), provide both socializing and reinforcing organizational messages about norms,
values, and expectations conceming research (Kuh & Whitt, 1988). Based upon Clark’s
perception of faculty as embracing multiple organizational affiliations (Clark, 1987),
Austin (1992) identified four faculty cultures: the discipline, the employing institution
(i.e., the university or college), the national system of higher education, and the scholarly

profession.

Disciplinary cultures have been isolated as the “primary units of membership and
identification within the academic profession” (Clark, 1987) and have been categorized in
many different ways such as pure or applied, abstract or practical, humanistic or
scientific, among others (Becher, 1987; Biglan, 1973; Gaff & Wilson, 1971; Snow,

1959). According to Austin (1992), institutional cultures are formed by a colleges or
university’s “institutional mission and purpose, its size, complexity, age, and location, the
way in which its authority is conceived and structured, the organization of work
(especially teaching and inquiry), the curricular structure and academic standards, student

and faculty characteristics, and the physical environment (p. 1617).” Clark defined

national systems of higher education to vary by the held values concerning issues of

42

accessibility (universality) of higher education, specialization of faculty work, the
relationship between general and vocational education, and the balance of research and
teaching activities (Clark, 1983). Finally, the culture of the academic profession includes
a set of primary academic values such as intellectual inquiry and understanding, societal
commitment, academic honesty and integrity, academic freedom, and faculty
collaboration toward a community or scholars (Austin, 1992). Austin believes that these
four cultures contain mixed and often conflicting messages concerning organizational
values, beliefs and norms for faculty work (Austin, 1990a, 1990b, 1992, 1994; Austin &

Gamson, 1983).

Although much has been written about the pervasiveness of faculty cultures in the
shaping of faculty work environments in general, little attention has been given to
medical faculty. The exception is Blackburn and Fox (1976). In their study of medical
faculty socialization, medical faculty (especially clinical medical faculty) do not seem to
identify well with the norms and values of the academic profession. Their follow-up
study in 1983, Blackburn and Fox found that in all medical specialties the opportunity to
engage in research was not valued equally across various physician career stages.
Blackburn and Fox also found that the overall value physicians placed on academic
matters was extremely low in comparison to other non-academic values. The limited
evidence available from these two studies suggests that the culture of the academic
profession may not be a useful construct when examining medical faculty culture--
particularly when trying to isolate the influences exerted upon medical faculty’s research

productivity.

43

Across all of Austin’s four faculty cultures, the disciplinary culture may be the most
influential upon faculty research productivity. The literature in higher and medical
education clearly shows that faculty research productivity varies between, and often times
within, differing disciplines (Blackburn et a1., 1978; Boyer, 1990; Braxton & Bayer,
1986; James Steven Fairweather, 1996; Finkelstein, 1984; Finkelstein et a1., 1998; Fulton
& Trow, 1974; McGee & Ford, 1987; Wanner, Lewis, & Gregorio, 1981). According to
F inklestein (1984), the most research productive disciplines are the natural sciences with
the “soft sciences” such as humanities, education, and art being the least research
productive--a fact which has been substantiated in many other research studies (Biglan,

1973; James Steven Fairweather, 1996; F inkelstein et a1., 1998).

As posited by Collins (1993), these differences reflect differing research philosophies and
access to research opportunities. These philosophies and research opportunities develop
into discipline-defining patterns of knowledge, ranging from “high internal consensus
patterns” in the sciences, to “low internal consensus patterns” in the humanities. These
consensus patterns are based upon the level of agreement about research foci and the
important universality of “truths” within different disciplines (Biglan, 1973; Braxton &
Bayer, 1986; Creamer, 1998). In addition, disciplines differ in both the number of

publishing venues and peer-review rigor (Finkelstein, 1984).

44

Workload, Rewards, and Faculty Research Productivity

Traditionally, faculty have been thought to narrowly divide their work time among the
three often conflicting roles of teaching, research, and service (Krahenbuhl, 1998).
However, substantial differences in this balance between time spent teaching and on
research are evident based upon a faculty member’s employing institutional type, with
research university faculty spending the greatest amount of time on research (1996). Prior
to World War II, teaching comprised the largest faculty role, but after the war due to an
explosion in federal research grant funding, teaching was soon replaced by research as
most dominant in faculty reward structures as manifested through formalized promotion

and tenure systems (Middaugh, 2001).

In terms of workload, studies in higher and medical education have shown that faculty
that include significant time devoted to research activities are more research productive
than those whose workloads include little or no time for research (Allison & Stewart,
1974; Bland & Schmitz, 1986; Calligaro & Others, 1991; Kraemer & Lyons, 1989;
Liddle, Westergren, & Duke, 1997; S. Vardan, H. Smulyan, S. Mookherjee, & K.
Mehrotra, 1990). Pelz and Andrews (1966) found that faculty workloads that are
balanced between research and other faculty roles, such as teaching and administration,
lead to greater productivity than exceedingly large or small workloads devoted to
research. A later study in university and industrial settings by Knorr, Mittenneir,
Aichholzer, and Waller (1979) showed that the most productive researchers are those

whose workloads included no less than 20%, nor more than 80% of time devoted to

45

research. The same study showed the ideal percentage of faculty time on research to

facilitate research productivity was an approximate 40% time commitment to research.

Academic institutions and academic departments greatly influence faculty research
productivity through their norms regarding faculty work (i.e., distribution of faculty
time), as well as reinforcement for conducting various faculty roles (i.e. intrinsic and
extrinsic rewards). The cultures of these academic institutions and departments exerts
influence on faculty through the establishment of research missions, various policies and
practices regarding research, organizational structures, and management and allocation of
research resources (Collins, 1993; Creswell, 1985a; Finkelstein, 1984; Kuh & Whitt,
1988). Further, faculty workload assignments, research budgets, salaries and wages,
promotion and tenure policies, and level of research leadership/mentorship as determined
by academic institutions and departments have also been shown to strongly affect faculty
research productivity (Dill, 1986). Diamond (1993) believes that these institutional
influences direct faculty’s interests away from teaching, toward research as a primary

activity.

Faculty rewards, such as compensation or promotion and tenure, also has also been
shown to strongly influence faculty members interest in producing research, and in
limited research, has been shown to positively influence faculty research productivity
(Bland & Ruffin, 1992; Dill, 1986; Fairweather, 1996; Kasten, 1984; Oeffrnger, ROaten,
Ader, & Buchanan, 1997 ). Fairweather and Rhoads (1995) categorized faculty reward

systems, in tandem with faculty workload allocation, as “administrative action.”

46

According to F airweather and Rhoads, administrative action is comprised of faculty
workload allocation and faculty rewards, and when combined with current socialization
and self—motivation, are keys to faculty behaviors, such as time allocated to both

instructional and research activities.

Academic Prestige and F aculty Research Productivity

Finally, academic institutions and departments also affect faculty research through the
level of prestige or reputation they possess for research. This prestige in turn provides a
level of cumulative advantage not unlike that which highly experienced individual
researchers develop through their prior research training (Fox, 1983; Merton, 1973).
Blackburn, Behmeyer, and Hall (1978) found in their discipline-spanning productivity
study that researchers affiliated with high prestige institutions had higher productivity
than those affiliated with lower prestige institutions, a pattern found similar by other
studies such as McGee and Ford (1987), Crane (1965), Long and McGinnis (1981).
Additionally, having high research producing colleagues in academic departments
positively influences faculty research productivity (Bean, 1982; Fox, 1983; Glueck &
Jauch, 1975). Maintaining conversations about research within and outside departments
and institutions has been shown to influence faculty research productivity by Blackburn,
Behmeyer and Hall (1978), Braxton (1983), and Pelz and Andrews (1966), and

Finklestein (1982).

47

Research Environments and Faculty Research ProductiviQ

Finally, Bland and Ruffrn (1992) provide a synthesis of the organizational literature
describing the relationship between research environments and productive faculty
researchers. Bland and Ruffin describe twelve important organizational variables (or
cultural characteristics) that positively influence faculty research productivity. These
cultural characteristics include organizations that have clear goals (including well-
developed linkages to individual’s goals), a research emphasis, a distinctive research
culture, a climate balancing between respect and “intellectual jostling”, assertive
participative governance, and a flat (decentralized) organizational structure. Further,
Bland and Ruffin include frequent communication, accessible resources (particularly
human), diversity of research groups (group size, age, sex, ethnicity, etc.), reward
structure for research, a focus on recruitment and selection, and leadership by those with

research expertise, as being keys to successful research productivity.

Toward a predictive model of family medicine department
faculty research productivity

Several literature-based and empirical models of faculty research productivity combine
the aforementioned factors in one form or another. Three of these models are particularly
useful because of the breadth of faculty productivity literature examined across multiple
disciplines. The first model by F inkelstein (1984) is based upon seven “normative and
behavioral variables” predictive of faculty publication productivity. These predictors
include faculty researchers having a research orientation, the highest terminal degree

within a field, early publication habits, previous publication activity, communication with

43

disciplinary colleagues, being subscribed to a large number of journals, and having

sufficient time allocation to research among other academic role components.

The second model by Creswell (1985a) describes several correlates defining the
productive faculty researcher. Creswell found psychological and individual correlates less
explanatory than sociological (such as preference for research, time spent on research,
and early productivity) and work-environment related correlates (such as prestige of

employing institution).

According to Creswell, the profile of the productive faculty researcher includes being
employed in a major university that rewards research and assigns ample time for faculty
to conduct research. Additionally, successful researchers tend to hold a senior
professorial rank, spend at least one-third of their time on research activities, publish
early in a career and receive positive feedback from peers for research efforts. Finally,
Creswell believes that successful faculty researchers maintain regular and close contact

with colleagues on and off campus who conduct research on similar topics.

The third faculty research productivity model used here was developed by Bland and
Schmitz (1986). Focusing on medical education, Bland describes successful faculty
researchers as having a personal motivation to produce research, an in-depth knowledge
of a research area, basic and applicable research skills, prior socialization to research, and
research mentorship. Additionally, Bland states that having an early record of

scholarship, well-developed professional networks, productive local peer support for

49

research, multiple research projects underway, and sufficient work time devoted to
research are also defining characteristics of successfirl faculty researchers. Finally, Bland
believes that successful faculty researchers have an external and internal orientation to
colleagues producing research, a high degree of autonomy in faculty work and in work
goals, and supportive environments for research that includes adequate physical and

human resources with institutions and departments.

For this study, I developed a conceptual model of family medicine faculty research

productivity based on the work by Finklestein, Creswell, and Bland (see Figure 2.1).

50

Figure 2.1

Conceptual Model:

Predictors of Family Medicine Department Faculty Research Productivity

 

 

 

 

 

 

 

 

 

 

l

 

 

 

Cumulative Advantage Individual Faculty Research Environments
(Prior Research Characteristics
Socialization)
# r
. Prestige of medical Demographic Disciplinary environment
school characteristics for research
Prestige of residency 0 Age 0 Professional networks
Prestige of graduate 0 Gender 0 Disciplinary norms &
school . Ethnicity expectations for research
0 Prior mentorship for 0 Advanced degree(s) . . .
research . Academic rank Instrtutgnal envrrlpnment
0 Prior fellowship/post- . Tenure status . r researc ,
doctoral training (with . . O Prestige of Institution
0 Tune In a faculty . .
research core) . 0 Institutional norms &
apporntment .
Prior research skills . expectations for research
P _ h . 0 Tune spent on
nor researc project research Departmental environment
experience
, for research
' Prior research Psychological & cognitive . Prestige of department
productrvrty: 1011111811s characteristics
. Pr' h _ . Departmental norms &
101' researc ' Motivation to do expectations for research
PTOdUCthlD’3 research . .
conference 0 Mentorshrp & leadership
presentations 0 Degree of role stress for research
0 Research agenda . Research support
0 Clear expectations for (resources)
promotion and/or 0 Protected time for
tenure research
' In depth knowledge 0f 0 Productive colleagues
research area of
specialty
0 Research training
0 Multiple research
projects

 

AL

 

Faculty Research Productivity

 

Peer-reviewed
publications

Peer-reviewed presentations

National Grants

 

0 Journal articles
submitted and accepted

 

 

 

0 National conference
proposals/papers accepted

 

 

- Govt. grants
submitted/funded
0 Private grants
submitted/funded

 

51

 

This model also incorporates the faculty cultural elements suggested by Austin that shape

and define faculty work, particularly research production.

The first section of the conceptual model includes the independent variables descriptive
of cumulative advantage (or prior research socialization). These variables include the
research socializing experiences as well as the research experiences future family
medicine faculty have had prior to the first faculty appointment. These include critical
research events undertaken during medical school, residency, graduate school, and
fellowship training. The independent variables in this section include research prestige
(reputation) of medical school, residency, and graduate schools (if applicable),
mentorship for research, fellowship/post-doctoral training, as well as prior research skills,
research project experience, research productivity, including journal articles written and

conference presentations given.

The second section of the conceptual framework, individual faculty characteristics,
contains two subsections of independent variables: demographic and psychological
characteristics of faculty. Both subsections involve current (i.e., since becoming faculty
members) rather than prior characteristics. The demographic subsection of independent
variables provide a description of faculty’s age, gender, ethnicity, advanced degrees
earned, academic rank, tenure status, time in faculty appointment, and time spent on
research activities. The psychological and cognitive subsection contains knowledge,

skills, and attitudes (including motivation) concerning research that may influence

52

research productivity. The independent variables in this subsection include having
motivation to do research, possessing an amount of role stress, having a defined research
agenda, clear research expectations for promotion and tenure, an in-depth knowledge of a
research area of specialty, additional research training, and multiple research project

underway.

The third section of the conceptual framework includes those independent variables
comprising family medicine department faculty’s research environments, including
subsections for faculty’s disciplinary, institutional, and department environments for
research. The independent variables in each of these three subsections include
organizational cultural elements that may be influential upon the current research
socialization of family medicine department faculty. The disciplinary environmental
variables include maintaining professional networks and the perception of family
medicine as a discipline rooted in research. The prestige of the institution and
institutional norms and expectations for research make up the institutional environment
for research. Finally, the departmental environment for research includes prestige for
research, departmental norms and expectations for research, mentorship and leadership
for research, research support (resources), protected time for research, and productive

colleagues.

The fourth section of the conceptual framework comprises the dependent (outcome)

variables for this study. These variables describes the research productivity of family

medicine department faculty in terms of peer reviewed scholarship, including the

53

traditional research productivity categories of journal articles submitted and accepted,

national conference presentations accepted, and national government and private grants

submitted and funded.

In the next chapter, the methodology for this study is described that involves gathering
descriptive data addressing each of the predictors from the conceptual model. Further, the
methodology section includes a description of how the conceptual model will be further
validated and tested empirically. Finally, the section will include a description of the data
collection tool for this study--a national survey of family medicine department faculty’s

research activities and interests.

54

CHAPTER III

METHODOLOGY

This study’s focus is to describe the overall research productivity of US. family medicine
faculty and the various factors that may be associated with research productivity.

Specifically, this study addresses the following questions:

(1) What are the individual faculty characteristics (demographic, psychological &

cognitive) of family medicine department faculty?

(2) What prior socializing experiences in terms of research have family medicine

department faculty had before accepting their first faculty position?

(3) What are the characteristics of family medicine department faculty's research

environments?

(4) What is the research productivity of family medicine department faculty in US.

medical schools?

(5) What prior socializing experiences, individual faculty characteristics, and

characteristics of research environments of family medicine faculty are predictors of their

research productivity?

55

Consistent with the goal of making national estimates of research productivity, this study
gathers survey data from family medicine faculty nationwide. This research is the first
survey study of family medicine department faculty ever conducted with faculty on a
national basis to determine their self-reported research productivity, as well as some of
the factors that may contribute to their productivity. Further, this study complements
existing demographic descriptive data concenring family medicine department faculty
gathered on an ongoing basis by the Association of American Medical College’s

(AAMC) Faculty Roster System (FRS).

The first section of this chapter describes the study’s population within departments of
family medicine nationally. The following section conveys an accounting of this study’s
sample and how it was determined. The third section involves a review of the data
collection strategies employed in this study include instrumentation, description of study
variables, validity measures, and data collection procedures. The final section includes a
detailing of the data analysis plan including attention to returns and non-returns, the
methods for handling response bias, the reporting and scaling of independent and
dependent variables, and the statistics used to compare variables and answer the research

questions.

56

Population
For more than 30 years, the AAMC’s FRS, in cooperation with the National Institutes of
Health, has collected census data on faculty in US medical schools across specialties and
academic departments. As of December 31, 1999, Table 3.1 shows the population of
family medicine faculty by advanced degree for the departments of family medicine
located at the 125 US. medical schools and geographically separated campuses (i.e.,

including US. territories).

57

Table 3.1

Population of US. Medical School Departments of Famm Medicine Faculty (by

Advanced Academic Degree)

 

 

 

Advanced Degree Number of Faculty % of all Family Medicine
Department Faculty

M.D. only 1,765 61%

MD. plus other degrees 324 1 1%

DO. only 88 3%

DC. plus other degrees 8 <1%

All other degrees and 696 24%

combinations (not included

in this study)

Total Family Medicine 2,881 100%

Faculty in US. Medical

Schools

58

For this study, the population includes all family medicine department faculty in US.
medical schools holding the MD. (61% of all family medicine department faculty) or
DC. degree only (3% of all faculty), and holding either the MD. or DC. degree in
combination with other advanced degrees (11% and <1% respectively). This population
includes 76% of all family medicine department faculty nationally. The 24% not
included in this study's population hold a wide range of advanced degrees, including the
Ph.D. degree and various masters degrees such as the Masters of Public Health, Masters

of Business Administration, and Masters of Education degrees.

Study Sample
A stratified random sample was selected for faculty holding the MD. only and the MD.
plus other degrees. For the MD. degree only and MD. plus other degree groups, the
Association of American Medical College’s Faculty Roster System drew random
samples. Both of these randonrly sampled groups cut across all departments of family
medicine faculty nationally. A certainty sample was used for two other groups--those
holding the DO. degree only and those holding the DO. degree plus another advanced

degree—because of their relatively small numbers.

A sample size of 3 14 for the MD. degree only group was chosen based on the formula

suggested by Aday (1989):

11 (minimum sample size)= (Standard error associated with a confidence leve12)(Estimate

of the standard deviation2)/desired precision), or n=(l .962)(2.02)/.05=314.

59

Both the standard error associated with a confidence interval (1.96) and desired precision
figures (.05) were arbitrarily set by the researcher using conventional figures used in a
majority of studies in medical and higher education. The standard deviation (SD) for this
study (2.0) was estimated by taking the average SD from a study conducted by
Hekelman, Zyzanski, and Flocke (1995), which was similar in methodology, design, and
focus. This study showed standard deviations of means ranging between 1.4 and 2.4; 2.0

was used as an arbitrary midpoint estimate of the standard deviation for this study.

Aday then suggests making several adjustments to the minimum sample size according to
the nature of the study design, expected response rate, and expected proportion of eligible
participants based upon the study’s inclusion criterion. Because this study involved a
random sample no adjustment in sample size is needed based upon the design of the
study. However, an adj ustrnent in the sample size must be made for the expected
response rate for this study. Using as a range the Hekelman et. al study’s (64% response
rate) and an arbitrary 50% response rate on the low end, 57% was selected as an
estimated midpoint target response rate. Using Aday's adjustment for the expected
response rate (minimum sample size/expected response rate, or 3 14/ .57), 551 was the
adjusted minimal sample size. Finally, Aday recommends an additional sample size
adjustment based upon the percentage of eligible responses. Since this study used FRS
data reportedly 90% accurate (by AAMC estimates), I divided the minimum sample size
by the percentage of eligibles (55 1/ .90), resulting in a final adjusted sample size of 612. I

rounded the MD. only sample size to 600 for convenience.

60

Using the ratio of estimated sample size to the population estimate for the MD. only
group, I selected 100 from the MD. degree plus other degree group. To this total I
included the identified national population of DC. only and DD. plus other degrees (88

and 8, respectively).

61

Table 3.2

Famin Medicine Dgiartment Faculty Stug/ Sample Size (by Advanced Academic

 

 

 

Degree)
Advanced Degree This Study’s % of Overall % of All Family
Sample of Faculty Study Sample Medicine
(by Academic Faculty
Degree) Nationally
M.D. only 600 75% 34%
MD. plus other degrees 100 13% 31%
DC. only 88 11% 100%
DC. plus other degrees 8 1% 100%
Total Sample Size 796 100% 36%

62

Data Collection Strategies
Instrumentation
I developed a survey instrument corresponding to the four parts of the conceptual
framework (see Appendix B for the survey instrument). The survey instrument also
included a variety of items derived from the higher and medical education literature on
faculty research productivity. 1 pre-tested the instrument with two national experts on
faculty research productivity, Dr. Carole Bland from the University of Minnesota and Dr.

John Creswell from the University of Nebraska.

Part one of the survey contained one question for each of the prior research training
variables, part two included one question for each of the current research interests and
activities variables, and part three contained one question for each of the current research
environment variables. In part four of the survey, one question was asked for each of the
three outcome productivity measures for this study (self-reports of peer reviewed
manuscripts for journal publication, peer reviewed proposals/papers for national
conference presentations, and national government and private grants) over the most
recent two-year period. Additional questions were asked to distinguish between
submitted and accepted manuscripts, as well as national government and private grants

submitted and actually partially or fully funded.

The prior research training, current research interests and activities, and current research
environment sections of the survey instrument used five point Likert scales that included

9, 6‘ ’9 ‘6

“strongly disagree”, “disagree , neither agree/disagree , agree”, “strongly agree” points

63

as suggested by Fink (1995) for scaled survey instruments. A sixth scaled option was
added called “not applicable” based upon the results of the family medicine focus group
feedback on the survey instrument (see validity measures section of this chapter).
However, responses using this sixth point on the scaling were not included when
tabulating the descriptive data for this study to allow for continuous measures to be
realized for each question. The self-reported research productivity section include a five
point scale ranging from zero to five or greater units of research productivity. This scale
emphasized lower levels of research productivity based upon relatively low anecdotal
estimates of family medicine department previously identified in the family medicine
literature for each productivity measure. Finally, the demographic section was primarily
composed of forced choice categorical measures, but also included continuous measures

for year of birth and hours per week spent on research activities.

Description of Study Variables

 

The conceptual framework and instrumentation for this study included several categories
of independent and dependent variables. The prior research socialization category of
independent variables focused on research and research-related skills and experiences
that family medicine faculty had prior to entering their first faculty position, primarily
during medical school and residency. Individual faculty characteristics included
independent variables that may be influential upon faculty research productivity, such as
age, gender, ethnicity, and tenure status, among others. Additionally, individual faculty
characteristics included a set of psychological and cognitive characteristics such as

motivation to do research, degree of role stress, and having a formalized research agenda,

64

among others. Research environments contained independent variables at the
disciplinary, institutional, and departmental level including professional networks,
institutional and departmental prestige in terms of research as well as norms and

expectations to conduct research, among others.

Validity Measures

 

In addition to checking the instrument with Drs. Bland and Creswell, the survey was
piloted during January 2000 with four family medicine department faculty members at
Michigan State University. The pilot respondents were asked to complete the survey,
time themselves, and address the following reliability issues (Fink, 1995):

o Are the instructions for completing the survey clearly written?

0 Are the questions easy to understand?

0 Do respondents know how to indicate responses (e. g., circle or mark the response,
etc.)?

0 Are the response choices mutually exclusive?

0 Are the response choices exhaustive?

0 Do respondents understand what to do with completed questionnaires (i.e., how to
return them and where to return them?

0 Do respondents understand by when to return the completed survey?

Based upon the feedback from the pilot group, the survey’s instructions were simplified,

a “not applicable” category was added to the survey instrurnent’s scaling, and the

survey’s cover letter was shortened.

65

Data Collection Procedures

The Association of American Medical College's Faculty Roster System compiled a
mailing list and prepared sets of mailing labels for each of this study's four sampled
groups. The surveys were mailed to respondent’s academic departmental offices. Prior to
the second mailing wave of surveys, non-respondents from the Faculty Roster System
mailing list from wave one were cross-referenced with the American Academy of Family
Physicians National Database to determine a new preferred mailing address (usually
either a clinic address or a home address). The second and third waves of surveys mailed
were to the preferred address to increase the likelihood of return response. Surveys for
each mailing wave were printed on different colored paper and were coded with a number

corresponding to each potential respondent in the study.

Using Dillrnan’s (1978) total design method for survey research, the cover letter
accompanying the survey was co-authored by the President of the Association of
Departments of Family Medicine (ADF M) on ADFM stationary to lend name credibility
(see Appendix A). Additionally, the American Academy of Family Physicians

Foundation (AAFP/F) supported production and mailing costs for this study.

The instrument was formatted to fit on two pages (with print on both the front and back
side), creating the impression that the survey was time-efficient. Further, respondents
were provided with a postage-paid envelope in which to mail back completed surveys.

All respondents were guaranteed confidentiality. Finally, the cover letter described

66

participation as being voluntary, with return of the survey signifying agreement to

participate in the study.

The mailing and data collection methods employed for this study are those also suggested

by Dillman for mail surveys:

1. At one week, a postcard reminder was sent to all faculty in the study to serve as a
thank you for those who responded, as well as to serve as a reminder to those who
had not responded.

2. At three weeks, a letter and replacement survey was sent to all non-respondents.

3. At seven weeks, a final mailing similar to the one in week three was sent with a final

replacement survey.

Data Analysis
Data analysis included: (a) a focus on returns and non-returns, (b) methods for dealing
with response bias, (c) reporting of a descriptive analysis of all independent and
dependent variables, ((1) scaling of independent and dependent variables and (e) the
statistics used to compare groups or relate variables and answer the research questions
(Creswell, 1994). The data analysis section of this chapter concludes with a
summarization of the research questions, questionnaire items and associated analytic

strategies listed in Table 3.3.

67

Returns, Non-Returns and Response Bias

Table 4.1 shows both the return and non-return rates in percentage for each of the study
sample's groups. Table 4.2 tests for response bias, with a wave analysis for weeks two,
four, and eight by analyzing responses to two questions from each of the five sections of
survey instrument. I used these tests to determine whether or not differences between

early, middle, and late responders existed.

Descriptive Analysis of Independent and Dependent Variables

The descriptive analysis of the first three research questions included a reporting of the
mean, standard deviation, and range of scores for each variable, or in the case of
dichotomous or categorical variables, frequency distributions. The results are presented
separately by variable: (1) cumulative advantage, (2) demographic characteristics, (3)
psychological and cognitive characteristics, (4) disciplinary environment for research, (5)

institutional environment for research, and (6) departmental environment for research.

The fourth research question focused on the dependent measure, research productivity.
Again, the analysis included the mean, standard deviation, and range of scores. The
productivity data was presented in a cross-tabulated format by distinct productivity
measure (journal articles, national conference presentations, national government grants,
and private grants). These outcomes were also compared by age, advanced academic
degree, academic rank, and years of faculty experience. Appendices D-G include these

productivity tables listed by demographic variable.

68

Scaling of Independent and Dependent Variables

To reduce the independent measures to a manageable number, I carried out a factor
analysis with orthogonal rotation to create composite predictor variables each with its
own factor score (Babbie, 1990). A reliability analysis was conducted to determine the
internal consistency of each new composite or factor score. Each of the individual
variables was not grouped together prior to the factor analysis to allow each of the
variables to form new categories based upon their relationships with each other. The four
outcome measures were kept distinct to permit separate examination of the different types

of scholarly productivity.

Statistics Used to Compare Groups, Relate Variables, and Answer Research Questions

 

The fifth research question required regressing each form of scholarly productivity on the

factor scores and other demographics. A full model regression was used.

69

Table 3.3

Relationship Between Research Questions. Questionnaire Items, and Analytic Strategies

 

Research Question Number(s) Questionnaire Analytic Strategies

Number(s)
1: Prior Socializing Research 1-9 Descriptive Analysis for each
Experiences variable in this category of

independent variables.

2: Individual Faculty 10-16; 34-42 Same as above
Characteristics
3: Research Environments 17-26 Same as above
4: Faculty Research 27-33 Descriptive Analysis for each of
Productivity four research productivity measures
5: Predictors of Research N/A Factor Analysis to combine
Productivity independent variables within each

major category of variables into six
overall categories. Then, a multiple
regression analysis to determine
which of the six categories is
predictive of each of the forms of

research productivity.

70

CHAPTER IV

RESULTS

Introduction
This chapter contains the results of a national survey of family medicine department
faculty’s research productivity as well as the factors that may contribute toward their
research productivity. The first section of the chapter contains a description of a survey
response rate and survey question analysis both conducted as part of the validity and
reliability measures for this study. The second section of this chapter provides the
descriptive results garnered for the independent variables for this study, while the third
section is comprised of the descriptive findings for the dependent variables (i.e., the
research productivity data of family medicine department faculty). The fourth and final
section of this chapter conveys the results of a factor analysis conducted to reduce the
study’s independent variables, as well as full model multiple regression tests of this
study’s conceptual model to determine the factors predictive of family medicine

department faculty’s research productivity.

71

Validity and Reliability Measures
In addition to the pre-tests described in Chapter 3, reliability tests were carried out using

survey response rate analysis and survey question analysis.

Survey Response Rate Analysis

The first wave of 796 surveys was mailed during January 2000. Prior to a postcard
reminder mailed in February 2000, 171 surveys were returned for the first wave for a
response rate of 21.48%. After the postcard reminder, a second wave of surveys was
mailed during March 2000. Prior to the third wave of surveys sent, 428 had been received
for a response rate of 53.76%. The third and final wave of surveys was sent during April
2000. Data collection was concluded during May 2000 with 474 surveys having been

received for a final response rate of 59.55%.

A review of the initial sample of 796 showed that 44 were ineligible for the study. Of the
44, 26 had left academic medicine, departments of family medicine, or retired, and 18
had moved without providing a forwarding address. The final adjusted response rate for

this study, excluding these groups, was 63.03% (474/752).

Table 4.1 provides a survey response rate comparison between the study’s sample and the
survey respondents by advanced academic degrees held. There is little evidence of bias

shown through the analysis.

72

Table 4.1

Survey Respon_s_e (by Advanced Acagemic Degr:ees Held)

Percentage of Percentage of Survey
Advanced Academic Degree Sample Respondents
M.D. only 75% 68%
MD. plus other degrees 13% 18%
DO. only 11% 12%
DO. plus other degrees 1% 2%

Survey Question Analysis (by Survey Wave)
A survey question wave analysis compared responses from each of the first four sections
of the survey for waves one and waves two and three (see Table 4.2). Waves two and

three were combined to allow similar sized comparison groups.

73

Table 4.2

Survey Question Wave Analysis
Survey Question

Medical School Known for Research

(Prestige of Medical School)

Residency Known for Research (Prestige
of Residency)

Research Satisfying (Motivation to Do
Research)

Pressure to Engage in Research (Degree
of Role Stress)

Network Outside Department
(Professional Networks)

Family Medicine Research as Discipline
(Disc. Norms & Expect. For Research)

Manuscripts Submitted for Publication

Manuscripts Accepted for Publication

Wave
Number

1
2&3

l
2&3

1
2&3

1
2&3

1
2&3

l
2&3

1
2&3

l
2&3

Mean
Score

3.52
3.42

2.79
2.74

3.56
3.34

4.20
4.07

2.30
2.43

2.40
2.43

1.32
1.00

1.01
.83

t

.84

.43

1.93

1.27

—1.07

-.30

2.09

1.32

df

468

458

446

448

435

460

454

453

Sig.
(2-
tailed)
.40

.66

.06

.21

.29

.76

.04

.19

When comparing mean responses from wave one with mean responses from waves two

and three, the only marginally significant difference noted was in the factor “manuscripts

accepted for publication”. Perhaps the respondents in wave one had a higher interest (and

value) in research thus their earlier response. Similarly, respondents in wave one were

more research productive than those in waves two and three. In sum the wave analysis

demonstrates little bias in the survey responses.

74

Descriptive Findings: Independent Variables
Descriptive data for the independent variables comprise the next four subsections of this
chapter. These subsections include a description of the demographic characteristics of
family medicine faculty participating in the study, and their psychological and cognitive
characteristics, cumulative advantage (prior research socialization), and research

environments.

Demographic Characteristics

Table 4.3 contains a summarization of demographic data of family medicine faculty in
terms of age, years of faculty experience, and hours per week spent on research activities.
Faculty in the study are relatively middle aged and report a moderate amount of faculty
experience. The faculty also spend little time on research activities, with a large majority
spending little if any time on research. Although there were small pockets of faculty
spending a significant amount of time on research, 37.6% reported no time spent on
research each week, 79.3% spent a half-day or less per week on research, and 88.2%

reported that they spent a day or less a week on research related activities.

Table 4.3

Demographic Chgacteristics: Age, Years of Faculm Experience, and HOurs per Week
Spent on Research Activities

Mean Median Mode Std. Deviation Minimum Maximum

Age 45.92 45.00 44.00 8.32 31.00 76.00
Years of Faculty 10.88 9.00 7.00 7.18 1.00 40.00
Experience

Hours Per Week Spent 3.41 1.00 .00 6.30 .00 40.00

on Research Activities

75

As shown in Table 4.4, the faculty were also predominately male and overwhelmingly

from non-minority ethnic groups.

Table 4.4

Demoggaphics Characteristics: Gender and Ethnic Origin

Gender

Frequency Valid Percent
Male 331 70.7
Female 137 29.3
Total 468 100.0
Ethnic Origin

Frequency Valid Percent

White, not of Hispanic Origin 404 89.4
Black, not of Hispanic Origin 15 3.3
Asian or Pacific Islander 14 3.]
Puerto Rican (Hispanic) 8 1.8
Mexican American of Chicano (Hispanic) 5 1.1
Other Hispanic 4 .9
American Indian or Alaska Native 2 .4
Total 452 100.0

Table 4.5 provides summary data in terms of academic rank. 74% of faculty categorized
themselves as holding traditional academic ranks (i.e., instructor, assistant professor,
associate professor, or professor) while 19% listed clinical ranks. The most prevalent

level given was at the assistant or clinical assistant professor level (55% in total).

76

Table 4.5

Demographics Characteristics: Academic Ra_nk_

Academic Rank

Instructor

Assistant Professor
Associate Professor
Professor

Clinical Instructor

Clinical Assistant Professor
Clinical Associate Professor
Clinical Professor

Other

Total

Frequency
10
167
93
72
7
68
28

1 1
9
465

77

Valid Percent
2.2
35.9
20.0
15.5
1.5
14.6
6.0
2.4
1.9
100.0

As shown in Table 4.6, in terms of academic status, faculty largely were in non-tenure
track academic positions. Additionally, mirroring this study’s sample, 87.7% held the
MD. degree, while 12.3% held the DO. degree, with several faculty reporting non-

clinical advanced degrees in addition to the MD. or DC. clinical degree.

Table 4.6

Demoggaphics Characteristics: Academic Status and Non-Clinical Advanced Degrees

Academic Status

Frequency Valid Percent
Not in a Tenure Track Position 341 73.8
Tenured 75 16.2
On the Tenure Track 46 10.0
Total 462 100.0

Non-Clinical Advanced Degrees
Frequency Valid Percent

Masters Degree in Health Related Area 58 12.5
Other Masters Degree 24 5.2
Ph.D./Ed.D. 1 1 2.4
Masters Degree in Education Related Area 9 1.9
Other Health Doctorate 2 .4

78

Table 4.7 contains a breakdown of the academic ranks from respondents in the study.
73.6% of faculty reported themselves in the traditional academic ranks of instructor,
assistant professor, or associate professor. Conversely, 25.5% were divided into clinical
academic ranks, such as clinical instructor, clinical assistant professor, clinical associate

professor, and clinical professor.

 

Table 4.7
Academic Rank

Frequency Valid Percent
Instructor 10 2.2
Assistant Professor 167 35.9
Associate Professor 93 20.0
Professor 72 1 5 .5
Clinical Instructor 7 1.5
Clinical Assistant Professor 68 14.6
Clinical Associate Professor 28 6.0
Clinical Professor 11 2.4
Other 9 1.9
Total 465 100.0

79

Psychological and Cognitive Characteristics

Psychological and cognitive characteristics were measured by respondent’s
agreement/disagreement with statements concerning their motivation to do research,
pressure to conduct research, ownership of a well-defined research agenda, understanding
of expectations for promotion/tenure, in-depth knowledge of a research area of specialty,
research training since becoming a faculty member, and having multiple research projects
underway. For each scale, the higher the number the greater the agreement with 5 being
the highest possible response. A score above 3 indicates an overall positive response,
whereas less than 3 indicates a negative response. Not applicable scores were recoded

and excluded from calculations.

As shown in Table 4.8, on the positive Side, respondents found research to be personally
satisfying, understood the promotion and/or tenure guidelines of their departments, stayed
“up-to-date” on literature in their research areas, and further developed their research
skills since becoming faculty members. On the negative side, a majority of faculty in the
study described themselves as lacking a clear research agenda, having few research
projects underway suitable for publication, and possessing a high degree of pressure to

participate in research activities.

80

Table 4.8

Psychological and Cognitive Chflcteristics

Mean Std.
Deviation

Find Research Satisfying 3.48 1.19
Feel Pressure to Engage in Research 4.15 1.09
Have a Well-Defined Research Agenda 2.52 1.23
Understand Research Guidelines for Promotion and Tenure 3.76 1.07
Stay up-to-date on Literature in Research Area(s) 3.71 .96
Further Developed Research Skills (since becoming a faculty 3.53 1.29
member)

Have Multiple Projects Underway 2.57 1.40

Cumulative Advantage (Prior Research Socialization)

Respondents described experiencing little exposure to research during their academic and

clinical training before becoming faculty members (Table 4.9). Few had substantive

experiences engaging in research projects or in developing their own individual research

skills. AS a result, few had published original research and even fewer had opportunities

to disseminate research at national conferences. Respondents also reported little

mentorship in research prior to becoming faculty, including have Opportunities to

participate in research-based fellowships or post-doctoral programs. A few faculty in the

study who had participated in graduate training prior to becoming faculty described their

experiences positively.

81

Table 4.9

Cumulative Advantage (Prior Research Socialization)

Mean Std.
Deviation
Medical School Known for Research 3.48 1.26
Residency Known for Research 2.77 1.32
Graduate School Known for Research 3.79 1.25
Mentoring in Research Prior to Faculty Position 2.35 1.34
F ellowship/Post-Doc Research Training 2.08 1.40
Research Skills Prior to Faculty Position 2.11 1.29
Meaningful Research Projects Prior to Faculty 2.49 1.41
Journal Articles Published Prior to Faculty 2.36 1.45
Conference Presentations Given Prior to Faculty 1.98 1.33

Research Environments

Family medicine faculty may be influenced by a variety of environmental factors
including the prevalent disciplinary cultural norms and expectations concerning research,
their academic institution’s prioritization of research, and the extent that their own
academic departments engage in research activities. As shown in Table 4.10, family
medicine faculty describe the disciplinary culture in academic family medicine as not
heavily vested in research. The majority of family medicine faculty do not have well-
developed professional networks of colleagues outside their own academic departments
with whom they can discuss research. In sum, characterization of family medicine as a

discipline lacking research focus seems supported by the respondents in this study.

Respondents view their institutions as strongly influencing research norms and
expectations. A large majority of the faculty viewed their academic institution as
possessing a reputation for meaningful research and creating an expectation for faculty to

generate scholarly work. For most family medicine faculty in the study, the institutional

82

expectation is realized-- that to be a successful faculty member means contributing a

body of original research.

Study respondents were less clear about their department’s cultural norms and
expectations for research. Although departments emphasize research, respondents do not
view their departments as possessing a reputation for research, nor are their colleagues
research productive. Further, research support has been unevenly manifested in resources,

including a consistent block of protected time for research.

Table 4.10

Research Environments: Disciplinary, Institutional, and Departmental

Mean Std.
Deviation
Networks Outside Department 2.34 1.24
Family Medicine Research as Discipline 2.41 .96
Institution Known for Research 3.40 1.26
Institution Emphasizes Research 3.69 1.16
Department Known for Research 2.76 1.18
Department Emphasizes Research 3.11 1.11
Department Chairperson Supports Research 3.54 1.12
Adequate Department Resources 2.91 1.29
Department Provides Protected Time 2.20 1.17
Department Faculty are Productive Researchers 2.20 1.00

83

Descriptive Findings: Family Medicine Department Faculty Research Productivity
This section describes research productivity as measured by submitted and accepted peer
reviewed manuscripts for publication, proposals/papers for conference presentation, and
submitted and funded national government and private grants. These outcomes are

examined separately by demographic variables.

Two Year Mean Productivity

Table 4.11 presents family medicine department faculty’s research productivity for their
last two academic years. For each type of productivity, faculty reported significantly less
than one scholarly work per year. Manuscripts for publication were the largest category

of scholarship generated.

Table 4.11

Family Medicine Department Faculty Research Productivity (Means for Last Two
Academic Years)

Mean Standard

Deviation
Manuscripts Submitted for Publication 1.20 1.61
Manuscripts Accepted for Publication .95 1.43
Proposals/Papers Accepted for Conference Presentations .71 1.30
National Government Grant Proposals Submitted .61 1.08
National Government Grant Proposals Funded .39 .82
National Private Grant Proposals Submitted .37 .82
National Private Grant Proposals Funded .22 .58

The majority of faculty produced nothing over the two academic year reporting period.
A frequency distribution of research productivity for the two academic years showed that

58.2% reported no manuscripts accepted for publication, 67.3% no proposals/papers

84

accepted for conference presentations, 75.8% no national government proposals funded,

and 84.3% no national private grant proposals funded (see Appendix D).

Productivity Differences by Demographic Variables

By age, the 40-45 year age group produced the most proposals/papers accepted for
conference presentation (.970) and obtained more private grants (.303). The 46-50 year
Old group had the most manuscripts accepted for publication (1.20) and national
government grants funded. Across all productivity measures, the 30—39 year age group
was the least productive. An AN OVA conducted by age for each productivity measure
shows that a statistically significant difference by age group in manuscripts accepted for
publication (F=9.80, df=3,p<.001), proposals/papers accepted for conference presentation
(F=4.32, df=3, p<.001), and national government grant proposals funded (F=4.06, df=3,
p<.001). No significant differences were found for national private grants firnded (see

Appendix E for details).

Males produced more of the following research outputs than females: manuscripts
accepted for publication (1 .08/ .629), proposals/papers accepted for conference
presentations (.78/.53), national government grant proposals funded (.41/ .34), and
national private grant proposals funded (.25/ .17). However, only the mean difference for

manuscripts accepted (t=3.08, df=451, p<.001) was statistically significant.

Caucasians appeared to out produce non-Caucasians in two of the productivity means,

manuscripts accepted (1.02/31) (t=3.18,df=437, p<.001) and conference presentations

85

(.78/.26) (t=2.48, df438, p<.001). However, for government grants (.40/.39) and private

grants (.23/.20), non-Caucasians out-produced Caucasians marginally.

Respondents with both clinical degrees (MD. or DC) and other advanced degrees
produced more research than those with the MD. or the DC. degree. Additionally,
across three of the productivity measures, MD’s out-produced D.O.’s (see Appendix F).
An AN OVA conducted by advanced degree for each productivity measure showed
significant differences between those with M.D.’s only, those with D.O.’s only, and those
with the MD. or DC. and other degrees, across all of the productivity measures:
manuscripts (F =9.89, df=2, p<.001), conference presentations (F =12.74, df=2, p<.001),
government grants (F =9.58, df=2, p<.001), and private grants funded (F =5.76, df=2,
p<.001). However, a post-hoe, Tukey HSD comparison of means for each of these
productivity measures did not show statistically significant differences at the alpha =.05
level between M.D.’s and D.O’s. This test did Show statistically significant differences
for each of the productivity measures between those with M.D.’s degrees only or those
with DC. degrees only, and those that have the MD. or DC degree in tandem with other

advanced degrees.

Full professors produced the most research, followed by associate professors, and
assistant professors. Faculty with clinical appointments produced the least for each of the
productivity measures (see Appendix G). An ANOVA showed a significant difference
among the academic ranks across all productivity measures: manuscripts (F=31.40, df=3,

p<.001), conference presentations (F=13.92, df=3, p<.001), government grants (F=16.29,

86

df=3, p<.001), and private grants funded (F=8.172, df=3, p<.001). A post hoc Tukey
HSD test revealed for manuscripts, conference presentations, and government grants,
there were significant mean differences between assistant and associate professors,
assistant professors and professors, and associate professors and those with clinical ranks

(V=.05).

Tenured or tenure track faculty produced more research of all types than did those who
were neither tenured nor on the tenure track: manuscripts (1 .96/ .59), conference
presentations (1.50/.44), government grants (81/24), and private grants (.52/.11).
Independent samples t-tests for equality of means showed significant academic status
differences for each of the productivity measures: manuscripts (t=-9.82, df=446, p<.001),
conference presentations (t=-8.12, df=447, p<.001), government grants, (t=-6.79, df=446,

p<.001) and private grants funded (t=-6.74, df=442, p<.001).

Faculty with 16 or more years of experience had the highest mean number of manuscripts
(1.40) and government grants (.67). Faculty with 11-15 years of experience had the
highest number of mean conference presentations (1.16) and private grants (.38) (see
Appendix H). For all productivity measures, those with five years or less of experience
were the lowest producers. An ANOVA conducted by years as a faculty member for each
productivity measure shows a significant difference between the experience groups
across all productivity measures: manuscripts (F=14.91, df=3, p<.001), conference
presentations (F =6.81, df=3, p<.001), government grants (F=10.43, df=3, p<.001), and

private grants funded (F=5.284, df=3, p<.001). A Pearson Correlation shows a moderate

87

positive correlation between years as a faculty member and manuscripts (.293, p<.001),
conference presentations (.171 , p<.001), government grants (.256, p<.001), and private

grants (.103, p<.05).

Finally, 3 Pearson Correlation shows a strong positive relationship between time spent on
research and each of the productivity measures: manuscripts (r=.557, p<.001), conference
presentations (r=.506, p<.001), government grants (r=.463, p<.001), and private grants

(1:412, p<.001).

Predictors of Family Medicine Department
Faculty Research Productivity
This subsection describes the results of a factor analysis conducted to create composite
predictors. Using these predictors, Research Question 5 is examined by testing the model

in Figure 2.1.

Results of Factor Analysis of Independent Variables

To reduce the likelihood of multicollinearity in the regression analysis composite
variables were created by carrying out a principal components analysis. Table 4.12
contains the results of principal components factor analysis using varimax with Kaiser

normalization. Factor loadings of less than .30 are omitted from the table.

88

Table 4.12
Factor AnalysiJs of Independent Variables

Rotated Component Matrixa

Component

1 2 3 4 5 6
Prior Research Project Experience .882
Prior Research Skills .855
Prior Research Productivity: Conference .851
Presentations
Prior Mentorship for Research .819
Prior Research Productivity: Journal Articles .777
Prior F ellowship/Post-Doctoral Research Training .766
Prestige of Graduate School .459 .405 -.326
Research Training .872
Research Agenda .823
Motivation to Do Research .774
Multiple Research Projects .769
In-depth Knowledge of Research Area of Specialty .660 -.300
Professional Networks .655
Clear Expectations for Promotion & Tenure .512
Mentorship and Leadership for Research .427 .374 .394 .374
Institutional Norms and Expectations for Research .816
Departmental Norms and Expectations for .810
Research
Prestige of Institution .799
Prestige of Department .779
Productive Colleagues .677
Protected Time for Research .419 .730
Degree of Role Stress -.716
Research Support Resources .345 .654
Prestige of Medical School .817
Prestige of Residency .377
Disciplinary Norms & Expectations for Research .849

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with
Kaiser Normalization.

“ Rotation converged in 7 iterations.

89

Six components emerged from the analysis. The first composite describes prior research
training before the first faculty position. The six measures in the composite include prior
research project experience, prior research skills, prior conference presentations given,
prior research mentorship, prior journal articles accepted for publication, and prior
fellowship/post-doctoral research training. The Chronbach’s alpha is .90, suggesting a

high degree of internal consistency among the variables in the composite.

The second composite combines measures of psychological and cognitive predispositions
toward research. This composite includes developing research skills through formalized
research training, possessing a definable research agenda, having motivation to do
research, having multiple research project underway, having an in-depth knowledge of a
research area of specialty, having professional networks, and having clear expectations
for promotion and tenure. The Chronbach’s alpha of the composite is .88, also suggesting

high internal consistency.

The third new composite defines the current research environment. It includes
institutional norms and expectations for research, departmental norms and expectations
for research, research prestige of the department, research prestige of the institution, and
having research productive colleagues. The Chronbach’s alpha is .85, demonstrating high
internal consistency. The fourth composite defines resources available for faculty
research. It includes having a department chairperson supportive of research, protected
time for research, the degree of role stress, and research support resources. The

Chronbach’s alpha of this composite is .64. The fifth composite defines institutional

90

prestige. It includes the prestige of the medical school, residency, and graduate school.
The Chronbach’s alpha of this composite is .60. The sixth composite contains a single
variable, perceptions of family medicine as a research discipline. Table 4.13 summarizes

the composites and shows relevant descriptive statistics.

91

Table 4.13
Descriptive Statistics: Composite Variables

N Mean Std. Deviation

Prior Research Training 314 2.27 1.16
Psychological and Cognitive Characteristics 363 3.27 .89
Current Research Environments 441 3.04 .91
Resources for Research 362 2.62 .80
Prior Institutional Prestige Regarding Research 459 3.34 .78
Perceptions of Family Medicine as a Research Discipline 462 2.41 .96

Results of Multiple Regression Analysis

After the factor analysis reduced this study’s independent variable to six new composite
predictors, a series of seven full model regressions were carried out to test the
contribution of each composite predictor toward predicting each of the productivity
outcome measures for this study. Along with the new composite predictors, demographic
variables that were not subject to data reduction were also used in each of the

productivity models.

Specifically, each model included the new composite variables prior research training,
psychological/cognitive characteristics, current research environments, resources for
research, prestige regarding research, and perceptions of family medicine as a research
discipline. Additionally, the following demographic predictors were used in each of the
models: age, years as a faculty member, hours per week on research, gender, ethnicity,
advanced degree (M.D.), advanced degree (Ph.D./Ed.D), academic rank (assistant
professor), academic rank (associate professor), academic rank (clinical ranks), and
tenure stream status. The use of full model regression analysis required the exclusion of

two demographic variables from each model, Academic Rank (Full Professor) and

92

Advanced Degree (D.O.), because of the statistical need to exclude at least part of
dichotomous variables in each predictive model. The seven productivity models included
manuscripts submitted and accepted, conference proposals/papers accepted, and national

government and private grants submitted and funded.

For the outcome manuscripts accepted for publication, the full model regression
accounted for 62.4% of the variance (R2), suggesting a strong relationship between the
independent variables used in the model and the outcome variable manuscripts submitted
for publication. An ANOVA conducted as part of the multiple regression analysis
showed that there is a highly significant relationship between the predictive models
independent variables and manuscripts accepted for publication (F=18.527; df=17,207;
p<.001). An examination of the coefficients for the predictive model shows that the
composite predictor psychological and cognitive characteristics and the demographic
variable hours per week on research activities provided the greatest contributions to the
prediction, with betas of .405 and .281 prospectively. Additionally, being in a tenure
stream appointment provided further, but less predictive contribution, as evidenced by a
beta of .125. The psychological and cognitive (t=6.383; p<.001) and hours per week
(t=4.371; p<.001) predictors were highly statistically significant, while being in a tenure

system appointment was statistically significant, but less so (t=2.234; p=.027).

For the outcome conference papers/proposals accepted, the full model regression
accounted for 50.4% of the variance (R2), further suggesting a strong relationship for the

predictive model. An ANOVA showed a highly significant relationship between the

93

predictive models independent variables and conference papers/proposals accepted
(F=11.425; df=17,208; p<.001). As with the manuscripts accepted for publication
predictive model, the coefficients for the conference presentations model also identified
psychological and cognitive characteristics and hours per week on research activities as
providing the greatest contributions to the prediction, with betas of .384 and .233
prospectively. However, the demographic variable of age provided further, but a less
predictive negative contribution, as evidenced by a beta of -.138. The psychological and
cognitive composite (t=5.285; p<.001) was highly statistically significant, while the hours

per week (t=3.171; p=.002) and age (t=-2.065; p=.040) was less so.

The government grants funded model accounted for 43.7% of the variance (R2), also
suggesting a strong relationship. An ANOVA showed a highly significant relationship

(F =8.708; df=17,208; p<.001). The model identified hours per week on research activities
and years as a faculty member as providing the greatest contributions to the prediction,
with betas of .586 and .196 prospectively. Hours per week was highly statistically
significant (t=7.482; p<.001), while years as a faculty member (t=2.274; p=.024) was less

SO.

The private grants firnded model accounted for 31.3% of the variance (R2), also
suggesting a strong relationship. An ANOVA showed a highly significant relationship
(F=5.03 7; dfi17,205; p<.001). The model identified the composites current research
environment and psychological and cognitive factors, along with hours per week on

research activities, as adding most to the predictive model, with betas of .222, .242, and

94

.202 prospectively. Additionally, each was statistically significant at the at the <.05 level,
with t values of 3.098 (current research environment), 2.817 (psychological and cognitive

factors), and 2.319 (hours per week on research activities).

Between each of the submitted and accepted models (i.e., manuscripts, government
grants, and private grants) there was little difference in the variance explained (see Table
4.14 for a summary). Additionally, while the manuscripts submitted and manuscripts
accepted models had the same predictors and beta significant levels, there were Slight
differences in the government and private grant models. For government submitted,
current research environments and family medicine as a research discipline are additional
statistically significant predictors along with years as a faculty member and hours per
week spent on research activities. The private grants submitted model varied from the
private grants funded model only in the exclusion in the private grants submitted model
of the predictor hours per week on research activities. To facilitate comparison between
each of the six predictive models, Table 4.14 provides a summarization of each of the full
model regression statistics, while Table 4.15 provides a synthesis of the model’s betas

and levels of statistical significance.

95

 

 

 

 

 

 

 

Sov 83 83 59v 83 83 83 a
Sq: Sq: ”out won: ”out SN .9 ”8.: a
Sea EA ”on.” one: as: 82: 3.: a
2m. 2m. 5.. 8m. 3m. 4%. me. am
unease.
SEESR
32...... 33.545 3.56 .3535 32.83.
35:“ mug-GHQ mun—ah” mun—ea n—amén—P—QEOQRQ EHQOQO< “antenna—W
0wfl>mhmm 0u3>mhn~ uflOEEO>OmV HEOEEO>OMV QOEQLQHGOU afitunflflaz afltumflflflz

 

 

 

 

 

 

 

 

 

 

semesm .082 as Accesses secs: Ea: :5 a 388%

3.... £an

96

Eva Jeanna... racemes?
sova 350%? racism .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we. ”8. 08.- 89- ms. :3. .32. hoe-w 0589
N8.- 8? 5.. $8.- 5.- 33.- E..- 3256 has
NS. 25.- 89- 9e. m3.- So. So. Sweden .83.... ”Ham
2:.- ms.. 27 m8. 8o.- E- E: eases as... ”Ram
8o. ”8.- a8. ”8.. R s. ”8.- N3; nan-5.5 ”ounce 805%...
3o. o8. 2 a. E. Re. as. as. a: ”canoe @8535.
5.- as. N8; 03. N8. me. as. bees-a
to. NS- as.- So; 48.- «8.. Go- EEO
:SN. m3. .3%. $2.. tam. .. A an. .. SN. fiaaom .5 EB am can
48.- m3 tee. ten. 2:. mg. as. cones: been a man;
08.. $oc 2 T :3.- 1a: .- e3.- 8? am...
usage
us.- 2o.- 30: :87 Re- N8.- 2? a a fleece 05282 been
2: .- 5.. So. So. us. as. 08. fiaaom magnum swans
m8. 2:. ”8. So. Re. to. 8o.- fiaaom ac masoam
:NNN. .12. So. rad. 48. So. as. afiaaesam fleece SEED
390mm
imam. awn. $0.- 3o. .33. .84. .84. 3255 2a acmeofiaa
2a.- 08. 48.- 3o. 2o. ms. may mean; Eases SE
"EDS Eeomofieeb
Engages SE: a:
82:5 855% 82:5 BEES .5383.
85:0 8520 $52G 8:26 23.39..— eoumoood. vast—55w
032.5 2.3:.— .rEU €50 oonououueo 33.535“: wanton—.5“:

 

 

 

 

 

 

 

 

 

 

3:395er tong-.355 35:8.— om 2 £52 .252 ii me afifiam

 

m y .v 2an

97

Summary
The results of a national survey of family medicine department faculty provide a
demographic profile. The typical family medicine department faculty member is male,
middle aged, Caucasian, not in the tenure stream, holding a traditional academic rank. In
terms of time spent on research, 79.3% of faculty reported spending a half-day or less per

week on research.

The results also provide a description of the psychological and cognitive characteristics
of family medicine faculty concerning research. Positively, faculty described a high
degree of internal satisfaction from research, have taken steps to improve their research
skills, and stay up-to-date in potential areas of research. Negatively, a majority of faculty
described a considerable amount of pressure to produce research, few had well-defined

research agendas, nor multiple research projects underway.

The results also provide a description of the prior research socialization before the first
faculty position. Before becoming faculty, few family physicians in training had
opportunities to conduct and disseminate research through publications and presentations.
Yet, the future family medicine faculty were largely aware of research activities having

trained at medical schools where research was emphasized.

In terms of the disciplinary, institutional, and departmental environments for research, the

influences of each are powerful, yet often times in conflict. The academic discipline of

family medicine is not viewed as a discipline well known for its research, yet academic

98

institutions are both known for and emphasizing research. At the departmental level,
Chairpersons are supportive of research, but there is not a strongly perceived emphasis
upon research. Further, departments are not perceived as being particularly well known
for research activities. Large impediments toward producing research described by
faculty include a lack of protected time and a dearth of departmental colleagues

producing research.

The results also provide insight on the research productivity of family medicine faculty.
In each of the faculty research productivity areas (manuscripts, conference presentations
and government and private grants), faculty reported significantly less than one unit of
productivity per year, with manuscripts being the largest category. An overwhelmingly
large majority of the family medicine faculty reported producing no research over the last
two academic years: 58.2% reported having no manuscripts accepted for publication,

67 .3% no proposals/papers accepted for conference presentations, 75.8% no national

government proposals funded, and 84.3% no national private grant proposals funded.

When examining productivity in light of demographic variables, a profile of the highly
productive family medicine department faculty emerges. Those that are relatively higher
producing tend to be Caucasian, male, between 40-50 years of age, and hold a DO. or
M.D. degree in tandem with other advanced degrees. Additionally, the higher producing
are more likely to be in the tenure stream, have eleven or more years of experience as

faculty members, and spend more than a half of a day per week devoted to research.

99

The results of this study also included the testing of a conceptual model of family
medicine department faculty research productivity. Predictive factors include six
composite variables (prior research training, psychological and cognitive characteristics
concerning research, current research environments, resources for research, prestige
regarding research, and family medicine as a discipline known for research).
Additionally, the model also included the demographic variables of age, years as a
faculty member, hours per week spent on research, gender, ethnicity, advanced degree
(M.D.), advanced degree (Ph.D./Ed.D), academic rank (assistant professor), academic

rank (associate professor), academic rank (clinical ranks), and tenure stream status.

For manuscripts submitted and accepted, the psychological and cognitive characteristics
composite, along with the demographic variables of hours per week on research and
tenure status, provided the largest contribution to the predictive regression model. The
conference proposals accepted regression model also had psychological and cognitive
characteristics, along with age, and hours per week on research activities as high
contributors. The government grants submitted and funded models shared the highest _
elements of years as a faculty member and hours per week spent on research. However,
the government grants submitted model had additional high contributions from current
research environments and family medicine as a research discipline. The private grants
submitted and funded models shared the psychological and cognitive and current research
environments composites as most explanatory. Additionally, hours per week spent on

research activities also was largely contributing for private grants funded.

100

CHAPTER V

DISCUSSION AND CONCLUSIONS

Introduction
This national study of family medicine department faculty attempted to determine their
research productivity as well as the predictors of this research productivity. This chapter
provides a discussion of the descriptive findings from this study, including family
medicine faculty’s individual characteristics, prior socializing experiences, current
research environments, and research productivity. This chapter also contains discussion
on the predictive factors and the conceptual model used as part of this study in light of
positioning this study’s results along side other faculty research productivity models, both
literature-based and empirical. Finally, by way of conclusion, the chapter ends with
specific recommendations for discipline of family medicine and the specialty of family
practice in light of this study’s findings toward fostering the growth advocated

throughout the profession.

Review of Descriptive and Predictive Findings

Family Medicine Department Faculty Demographics

The first section of this study provided a description of family medicine department
faculty’s demographic characteristics. Comparisons can be made between this study’s
demographics and data compiled by the Association of American Medical College’s

Faculty Roster (1999). The demographic profile of faculty that emerged from this study is

101

consistent with demographic data found in the Faculty Roster for family medicine

department faculty, in regard to gender, ethnicity, and tenure status.

However, the Faculty Roster data does not include separate breakdowns for clinical and
non-clinical academic ranks, nor does it specifically categorize those holding clinical
academic degrees in combination with other advanced degrees. This study found that
25% of family medicine faculty are located in clinical academic ranks and over 20% have
additional advanced degrees beyond the MD. or DC. degree, data points that are missed
through Faculty Roster reporting. Additionally, this study provides a unique contribution
by providing data on both faculty’s age as well as years in a faculty appointment, further

demographics not addressed by Faculty Roster data.

Perhaps the most interesting demographic finding that emerged from this study was that
79% of faculty spent less than a half-day per week on research and 36% spent no time at
all. These findings are consistent with earlier studies of family medicine faculty
conducted by Culpepper and Franks (1983), as well as Hueston (1993). Unfortunately,
the Faculty Roster does not measure faculty workload (including time spent on research),

making comparisons to this study’s workload data impossible.

However, a comparison can be made between this study’s faculty and mainstream higher
education faculty regarding time spent on research. The percentage of faculty in this
study spending little or no time on research is dramatically larger than reported for higher

education faculty by Fulton and Trow (1974) and Ladd (1979). Fulton and Trow reported

102

that, across all disciplines and institutional types, 7% of faculty spent no time on research
and an additional 15% spent between 1-4 hours per week on research. Ladd found similar
numbers in his study, 3% spending no time on research, and another 11% spending 1-4

hours per week.

The reasons for family medicine faculty spending less time on research than higher
education faculty may be two-fold. As identified in the family medicine literature, family
medicine faculty have been increasingly called to support academic activities through
increasing clinical responsibilities. Because of heightened clinical demands, there is
likely to be an erosion of the time available for research for faculty. Further, as shown in
this study, the perceived expectation by the discipline of family medicine, medical
schools, and departments of family medicine for faculty to conduct research has remained
low in comparison to higher education faculty, much less other medical specialty’s

faculty.

Psychological and Cognitive Characteristics

In addition to the demographic characteristics of family medicine faculty, this study also
provided some unique insights on the psychological and cognitive characteristics of
family medicine department faculty, particularly the tension between clinical and
academic responsibilities. Faculty in the study are aware of the need to produce research
and to develop their research skills toward becoming research productive scholars. Yet,

there is strong evidence from the results that faculty face a great deal of countervailing

103

pressure to be highly effective and profitable clinicians. Undoubtedly, this tension is

manifested in the extremely high level of role stress reported by faculty in this study.

In short, the tension between academic and faculty roles is also exacerbated by the mixed
messages being supplied by medical schools and departments of family medicine
regarding the relative importance of both of these faculty role domains. As with higher
education faculty possessing finite hours in which to complete multiple responsibilities,
family medicine faculty are forced to make choices about faculty workload composition
based upon their own strengths and preferences, but even more so, institutional and
departmental expectations manifested through faculty promotion and reward structures.
However, unlike higher education faculty, family medicine faculty must find ways to
balance not only teaching, research and administrative responsibilities, but also clinical
practice roles. This tension is supported by Blackbum’s belief that medical faculty
wrestle with two often conflicting professional identities, one as acaderrrics, and another
as a clinicians (Blackburn, 1976). Norrnatively however, the clinical identity emerges as
largely dominant in determining faculty’s overall beliefs, values, and attitudes, and

theoretically, prioritization of faculty work roles.

However, since family medicine faculty have chosen an academic career pathway, there
must be at least a moderate interest and appreciation for research. Studies by Fulton and
Trow (1974) and Ladd (1979) identified a large majority of faculty across disciplines and
institutional types as having at least a moderate level of interest in research activities. The

psychological and cognitive findings from this study in regard to interest and motivation

104

for research are also consistent with studies by Finkelstein, Seal and Schuster (1998)
showing the greatest difference between faculty’s preferred time and actual time spent in
all roles is in research activities. This provides evidence that family medicine faculty are
largely the same as higher education faculty in their intrinsic motivation for research, as

well as perceived stress in struggling to balance conflicting faculty roles.

Prior Research Socialization (Cumulative Advantage)

This study’s findings on prior research socialization shows little evidence of cumulative
advantage occurring prior to the first faculty position as is often found in higher
education, particularly in the sciences. Perhaps this is because of the extremely limited
opportunities in medical school, internship, and residency that could be considered actual
research socialization for future family medicine faculty, such as receiving research
mentorship, participating in research projects, and having experiences producing and
disseminating research. However, some future faculty were trained in medical schools
and graduate schools where research was emphasized that may have been somewhat

influential upon their overall development as researchers.

This lack of research socialization is largely different from what is experienced by
graduate students anticipating future faculty careers, as well for other medical specialties
students, interns, and residents developing their clinical practices rooted in a defined
knowledge base centered in disciplinary scholarship. Particularly for graduate students
during their academic training, there is significant exposure to high producing research

mentors that train them how to identify researchable questions, how to develop a research

105

agenda, and how to obtain resources for research. Because family medicine faculty
largely have missed this valuable anticipatory socialization, accelerated research
socialization as new faculty becomes vital toward assuming faculty responsibilities,

including producing research.

Environments for Research

Family medicine faculty members can be considered in some ways to be products of their
disciplinary, institutional, and departmental environments for research. As discussed
earlier, often conflicting and contradictory messages are sent to family medicine faculty
in regard to the norms and expectations for research that may differ at the disciplinary,

institutional, and departmental level.

From the results of this study, it is clear that family medicine faculty do not consider
family medicine as a specialty, nor family practice as a discipline, to be strongly rooted in
a research culture. Part of this perception may be because of the relative youthfulness of
both the specialty and discipline. But a larger reason may be what others have identified
as family practice’s “counter-cultural roots”, seeking to distance family practice from
other medical specialties through the creation of a unique populist clinical identity, as
well as non-conformist, “anti-intellectual” academic identity (Stephens, 1979; Rodnick,
1987; Doherty, 1987). It remains problematic for an emergent specialty and discipline to
have faculty who largely do not share a consensus for or against these emergent

professional identities.

106

The faculty in this study also describe little research networking occurring within
departments, medical schools, and among faculty as scholarly colleagues. One cause for
this lack of research networking may be because these networks are not developed much
less modeled in medical schools and residency programs through primary care research
activity and scholarly collaboration. Further, the lack of networking may occur because
few family medicine faculty attend national research-based conferences where many
linkages are forged both during their clinical development, as well as after being
appointed faculty. Finally, since family medicine faculty identify themselves primarily
as clinicians, their effort to maintain and develop networks may be focused more on

clinical practice networks than research ones.

Medical schools that faculty are located in may also play a significant part in influencing
research efforts because of these institution’s available research resources. It is clear that
most family medicine faculty are aware of the research norms and reputations of most of
these medical schools for clinical research production. But, as is the case in mainstream
higher education, it is doubtful that research expectations conveyed by academic
institutions carry greater weight than greater weight than research expectations exuded

from both disciplinary and departmental research environments.

The departmental environment for research for family medicine faculty seems to largely
mirror the disciplinary mixed messages that are delivered concerning research. On the
one hand, family medicine faculty describe their chairpersons as being supportive of

research and that most essential resources are available for research. On the other hand,

107

this encouragement and facilitation does not translate into a critical mass of faculty
conducting research, and as a result, few departments can be considered highly research
productive, nor do they tend to convey high expectations for research. Perhaps the
greatest department level impediment to research is that few departments offer protected
time for research--a benefit consistently mentioned in the higher education literature as

strongly influencing faculty research productivity.

Finally, there are several characteristics of family medicine department environments that
may make them less than optimal for research efforts. For one, because of their clinical
responsibilities, family medicine faculty may have a high degree of “academic
rootlessness” in that they often move from one clinical practice site to another several
times in one week, rarely setting foot in academic departments but for the occasional
faculty meeting. As a result, unlike in most higher education departments, a core of
family medicine faculty may have few opportunities to meet and collaborate on research
projects, much less to formally and informally discuss research interests and activities.
Finding ways to link faculty toward a common commitment to research (as well as other
faculty roles) remains one of the greatest challenges facing family medicine departments

nationally.

Family Medicine Department Faculty Research Productivity
Family medicine department faculty in this study reported very little evidence of
scholarly productivity. The fact that in a two-year period, 58% of faculty did not produce

any manuscripts, 67% any conference presentations, 75% any national government

108

grants, and 84% any private grants, clearly is troublesome to a specialty and discipline
desiring growth and definition through a critical mass of faculty and research. As with
time spent on research, the pattern of lack of research productivity is particularly more
pervasive than is found in mainstream higher education and other medical specialty’s
faculty and supports earlier productivity studies conducted in family medicine two
decades ago. Most troubling is the fact that this study offers no evidence that the level of
family medicine department faculty research productivity is on the increase even given

the consensus in the family medicine literature for the need for such scholarship to occur.

A Predictive Model of Family Medicine Department Faculty Research Productivity
The test of this study’s conceptual model is compelling because of the theoretical and
practical implications realized through this research. Perhaps most importantly, the
results of this test are particular helpful for family medicine leaders to begin formulating
discipline-wide research initiatives, and for medical schools and family medicine
departments toward initiating faculty development programs addressing elements of this

conceptual model.

The results of the test of the conceptual model for this study showed that a wide-range of
prior research socialization, psychological and cognitive factors of faculty toward
research, and research environmental variables are at the very least, moderately
influential upon family medicine faculty research productivity. However, in terms of a
predictive relationship, across most of the productivity measures, the psychological and

cognitive characteristics of family medicine faculty concerning research was most highly

109

predictive of family medicine research productivity. These factors include having further
developed research skills through formalized research training, having a defined research
agenda, possessing a motivation to do research, and having multiple research project
underway. Further, these factors also include having an in-depth knowledge of a research
area of specialty, well-developed professional networks, and clear expectations for

promotion and tenure.

It is important to note that unlike past higher education faculty research productivity
studies, the influence of the disciplinary, institutional and departmental research
environments is extremely small in comparison to the other factors examined in this
study. This may support the earlier argument made that due to mixed positive and
negative messages concerning research conveyed by family medicine as a discipline, as
well as medical schools and departments of family medicine in general, the overall level

of influence these environments have on productivity may be questionable.

Additionally, across all productivity measures, the impact of prior research training
(research socialization), institutional prestige factors regarding research, and the
availability of research resources appear to be largely negligible upon family medicine
faculty’s research productivity. This may be the case for research socialization because of
the wide lack of research exposure firture faculty receive during their clinical training,
making research socialization for the most part non-existent. Because so few departments
have a substantial critical mass of researchers, family medicine departments may be

noted for other prestige factors, namely clinical ones, which provide more reputational

110

value to departments. Finally, in terms of research resources, because there is not much
competition for the use of resources because so few faculty are conducting research

within departments, its value as a predictor is not surprisingly questionable.

As is the case with most other disciplinary studies of faculty research productivity, there
is little evidence from this study across all forms of research productivity, that the
demographic variables of gender, ethnicity, and academic rank are particularly predictive.
The lack of a demographic effect is not surprising given the relatively homogeneous
faculty composition in family medicine, as well as the widespread lack of research
activity cutting across all demographic groupings in family medicine. However, also
consistent with other faculty research productivity studies, the demographic variable
related to faculty workload-~hours per week on research activities--holds particularly
strong predictive strength across most of the productivity measures. Clearly having
faculty roles where research is expected and is a larger portion of the overall workload
allows faculty to devote concentrated time to research activities ostensibly less

encumbered from competing faculty roles.

Across all productivity measures there were some key differences in contributions from
various predictors. Research environments seemed to be more important for all forms of
grants, psychological and cognitive factors for government grants, and tenure status for
manuscripts. The influence of research environments for grants can perhaps be explained
because of the emergent influence of institutional pressure to address healthcare

financing, as well as the need to find alternative ways to support academic activities. The

111

tenure status relationship with manuscripts is not surprising given the current
expectations for promotion and tenure in medical schools centering on manuscripts as
predominant. However, the lack of a psychological and cognitive contribution to
government grants (while not for other forms of productivity) remains enigmatic, as does
the negative contribution of family medicine as a discipline for research toward

predicting government grants submitted.

Because few faculty research productivity studies, both empirical and literature based,
have comprehensively examined all of the elements of the conceptual used in this study,
comparisons need to be made between this study’s finding and these other studies by
examining individual factors. Returning to the three models of faculty research
productivity that serve as the foundations for the conceptual framework used in this

study, a number of important similarities and differences are noted.

Finkelstein (1984) stated that there are seven important factors that predict research
productivity across multiple disciplines. This study’s composite psychologiCal and
cognitive factors is in agreement with two of Finkelstein’s predictors: having
communication with disciplinary colleagues, and having a research orientation
(motivation). Clear differences also emerged in comparing this study’s findings with
Finkelstein’s. Two factors from Finkelstein’s study were largely predictive, but were not
supported here, including having early career publication habits and previous publication

activity. These differences may be accounted by the fact that as discussed earlier, the

112

research socialization of family medicine faculty may be largely different that what

occurs for mainstream higher education faculty in terms of research exposure.

In addition to Finkelstein’s model of predictive factors, Creswell’s model of faculty
research productivity described several correlates from the literature that are likely to
influence research productivity. These correlates include intelligence test scores,
motivation, personality characteristics, stress, age, gender, prestige of doctoral programs
and mentoring, prestige of employing institutions, research resources and assignments,
colleagues, rank and tenure status, early productivity, preferences for research, and
disciplinary differences. Of these, Creswell identified psychological correlates (such as
through intelligence tests) and demographic correlates (such as age, gender, and
ethnicity), as being less explanatory than sociological correlates, such as preference for
research, time spent on research, and early productivity--findings highly consistent with

this study.

Finally, the clearest linkage can be made between the predictive results from this study
and the comprehensive, literature-based model developed by Bland (1986). This is
because Bland’s model is comprehensive, cutting across many possible explanations for
research productivity, and because it was written to be applicable across multiple
disciplines. Of Bland’s thirteen “Characteristics of Successful Researchers”, eleven were
included in this study’s conceptual framework. Of these eleven, six characteristics were
supported through a majority of this study’s predictive outcomes. These six

characteristics include having a personal motivation for research, an in-depth knowledge

113

of a research area, basic and applied research skills, well developed professional
networks, multiple projects underway, and sufficient work time (protected time).
However, four of Bland’s characteristics were not supported as being predictive in this
study conceptual model. They included having research mentorship, early work habits,

productive local peer support, and a supportive research environment.

In sum, this study’s model of family medicine department faculty research productivity is
largely comparable to the few studies that examine research productivity from several
different explanatory categories of predictors. While there are elements of overlap and
agreement between this study’s predictive findings and other studies findings,
generalizations are problematic because disciplines often do not have similar socializing
experiences, institutional culture dynamics, and expectations for faculty research.
Ultimately however, the goodness of fit of this study’s conceptual model will be
determined by future researchers that attempt to use this framework in further faculty
research productivity studies. Additionally, an perhaps more importantly, the power of
this model will lie in the use that family medicine leaders and administrators have in
further developing family medicine department faculty researchers based upon the results
of this study. In the next section of this chapter, some recommendations are given for
these same family medicine leaders and administrators toward developing family

medicine researchers in light of the results of this study.

114

Recommendations for Family Practice and Family Medicine

The results of this study lend themselves to several recommendations for both the
specialty of family practice and the academic discipline of family medicine. These
recommendations have been suggested in the literature consistently by family physicians,
family medicine faculty, and administrators as being essential toward the development of
the specialty and academic discipline. However, before specific recommendations for
family medicine can be considered further, a new research culture in family practice
needs to be fostered by family physicians, both academically and clinically. This new
research culture is centered in a core or critical mass of family physicians being broadly

engaged in scholarly approaches to their clinical practice and in their faculty work.

Developing a research culture in family practice primarily involves convincing family
physicians to engage in multiple research roles within their clinical practice
environments. Leaders in family medicine are advocating that all family physicians need
to contribute to this new conceptualization of family practice as being grounded in
scholarly foundations. Stange et.al (2000) captured this most succinctly by stating that
“every individual and organization involved in practicing, teaching, administrating or
certifying general practice should participate in creating a culture that fosters the
generation of new knowledge.” Further, Stange and others in leadership positions in
family practice have stated that each family physician should engage in a self-reflective
practice that involves the patient voice in generating questions and interpreting clinical

data. In sum, according to Geyman (1978), “the single most important factor influencing

115

the extent to which family physicians can be involved in investigative work is the

‘mindset’ held by the physician toward his/her own practice (p.74).”

Problematically, cultural research conducted in higher education strongly suggests that
changing, or even influencing, the prevailing cultural norms and values in family
practice, and ultimately family medicine faculty’s research behavior patterns, will be
exceedingly difficult (Austin, 1983; Alasuutari, 1995; Chaffee, 1988; Kuh, 1988; Schein,
1992; Wagner, 1981). Yet, Peterson et. a1 (1986) identified seven ways in which cultures
in higher education may be influenced toward change. They include creating new
organizational units, changing clientele or staff, using visionary leadership styles,
redefining missions, reorganizing organizational units, using conflict creatively to
examine differences between examined and espoused values, and using key events and

conditions to refocus goals and priorities.

Once a change in the research culture has begun to take place, family medicine can begin
to consider longitudinal approaches that shift the research culture even further toward
family medicine department faculty producing original research. Using the research
frame for this study as a guide, in the next three sections of this chapter several specific
suggestions are offered for academic leaders in family medicine that may be useful for

influencing family medicine faculty’s research productivity.

116

Restructuring the Research Socialization of Future Family Physicians

and Potential Family Medicine Faculty

As evidenced by this study, as well as the few research studies that have been conducted
on family medicine faculty to-date, medical students, interns, and residents encounter few
socializing events that may later influence their decision to becoming academics and to
ultimately engage in research. However, several exposures to research activities could be
developed throughout their academic and clinic training toward motivating them to

embrace research early on.

During medical school, students interested in primary care (including family practice,
general internal medicine, among others) should be identified and strongly encouraged to
take a course on evidence-based clinical practices. This course should be developed and
taught by research productive primary care physicians from all primary care disciplines
who have successfully blended research into their own clinical practices. Ideally, this
course should focus on developing primary care research skills, formulating research
questions, as well as learning how to critically appraise the medical literature.
Additionally, as part of this course or a separate course, a capstone primary care,
discipline-specific research project should be undertaken. For example, with students
interested in family medicine, the generation of a capstone project could involve a family
physician being partnered with two or three medical students in a clinic setting to
examine the root and mediating psychosocial causes of patient non-compliance with
common drugs. At the end of this capstone course, family medicine students would

present their research to their peers within medical schools. Additionally, they would also

117

be encouraged to produce subsequent journal articles and present at national conference

toward disseminating their research.

Unfortunately, a majority of current didactics in medical schools focuses on systemic,
disease state fundamentals with little attention given toward providing opportunities for
faculty to engage in discussion of the results of their original research projects. This
problem in medical schools is especially acute for primary care specialties more than for
the subspecialties. To rectify this shortcoming, family physician researchers from both
the community and from medical school faculties need to be paired to develop dialogue
sessions with medical students on their own research projects and the implications of

their research for family practice as a specialty, and for society at large.

Additionally, both during medical school and during the internship year, career
development sessions should be developed that not only explore primary care specialty
selection, but also encourage primary care academic careers as options. These sessions
need to highlight equally the intellectual and other benefits of faculty careers, as well as
the normative expectations for research such faculty careers would entail. Further, the
career sessions should show that academic fellowships, with research as a core, have both
short-term and long-term clinical and academic benefits, and are viable and dynamic
alternatives from the traditional preparatory model moving from medical school,

internship, and residency directly to clinical practice positions.

118

Finally, once future family physicians have selected family practice as a specialty, their
residency programs need to encourage the continued development of research skills,
critical appraisal of the literature, as well as engagement in family medicine research
projects. While many family practice residencies currently offer these opportunities, they
are not viewed as longitudinal research projects, but instead are narrowly defined,
suitable for completion within a year or less. As a result, research projects lack the
breadth and depth that family practice research affords as its strengths. Clearly, family
practice residencies that span three or four years in length have the potential, if structured
early, of allowing faculty practice residents opportunities to develop the early career
research habits that have been proven to be predictive of higher education faculty’s

research productivity.

In sum, revising medical school and residency curricula is needed to allow family
medicine trainees more opportunities to interact with family medicine faculty researchers
as mentors, as well as to develop basic research skills, and engage in meaningful research
projects. As these socializing experiences coalesce toward the formation of a culture of
research, future family physicians will begin to be influenced by the norms and values for
research needed to become successful family physicians, and potentially, future family

medicine department faculty.

119

Drawing Strength from the Psychological and Cognitive Characteristics of Family
Medicine: The Recruitment, Selection and Retention of Faculty

As evidenced from the results of this research project, the psychological and cognitive
characteristics of family medicine department faculty are clearly the most important
factor that predicts their research productivity. However, many of the individual faculty
characteristics of faculty as well as their psychological and cognitive beliefs, values, and
attitudes about research are innate, and as such, are not easily subject to influence. Yet,
there are several practical, longitudinal steps that can be taken by administrators and
leaders seeking to shape family medicine faculty research productivity in regard to these

psychological and cognitive factors.

For one, the recruitment and selection process of new family medicine faculty by
institutions and departments needs to enhanced in several ways. Family medicine faculty
search committees need to be composed of research producing faculty and deans and
department chairs active in scholarship. Additionally, family medicine faculty job
descriptions need to more clearly define the research expectations required for promotion
and/or tenure to occur. Consideration needs to be given by search committees to not only
focus on clinical practice potential and teaching skills, but also interest and motivation to
do research. Search committees need to not only address candidates past research
productivity, but also pay more attention to their potential growth as researchers through
their ability to articulate a clear research agenda and ability to identify a cohort of fellow

primary care researchers with whom collaboration is possible.

120

Once new family medicine department faculty are hired into faculty positions,
comprehensive, yet individualized institutional and department orientation programs
should be offered to new hires. Meetings should be included in the orientation program
that allows faculty time with medical school deans and department chairs to discuss
establishing early (and ongoing) research expectations toward the development of
individualized research agendas. Medical school deans and family medicine department
chairs should actively engage in creating linkages for new faculty by formally scheduling
appointments with new faculty and existing productive researchers to provide them with

overviews of research activities currently undertaken both institutionally and nationally.

Finally, new family medicine department faculty should be required to develop an
individualized academic career committee that includes the family medicine department
chair, a senior family medicine faculty member, and another junior faculty member. This
peer committee would help new faculty determine suitable topics for research, identify
potential collaborative faculty researchers and research funding opportunities, and serve
as a sounding board for research ideas. These career committees should meet quarterly,
particularly early in a new faculty member’s career and would be gradually replaced by

informal meetings with departmental promotion and tenure committees if applicable.

121

A Reexamination of Family Medicine Faculty ’s Workload Policies (Roles and
Rewards)

Based upon the results of this study, leaders of national family medicine organizations
such as the Society of Teachers of Family Medicine, American Academy of Family
Physicians, and North American Primary Care Research Group need to jointly initiate
further dialogue about family medicine faculty’s research roles and rewards. Critical
workload issues need to be addressed through the establishment of faculty roles
conferences similar to the Faculty Roles and Rewards Forums held annually in higher
education by the American Association for Higher Education (AAHE). At these forums,
family medicine faculty leaders, administrators, and faculty need to develop policy

directives addressing some of the following questions:

I What are the various research roles family medicine faculty can play?

I Should all family medicine faculty be required to engage in some form of research?

I How should research be weighted, evaluated, and rewarded along with other faculty
roles?

I How can research activities be encouraged as part of daily clinical and academic
practices?

I What is the role of the tenure system for family medicine faculty in terms of research
expectations?

I How can family medicine faculty be supported in terms of balancing clinical and
academic roles?

I How can family medicine leaders and administrators encourage the development of

faculty research skills locally, regionally, and nationally?

122

Each of these policy issues need to be discussed in collaborative forums with discussions
then moving forward locally to medical schools and departments of family medicine to
reexamine promotion and tenure guidelines, recruitment practices, as well as targeted,
individualized faculty development. Through well thought out national and local research
policies and practices, family medicine can take large steps as an academic discipline in
encouraging faculty careers and achieving a critical mass of family medicine research and

researchers.

Revisiting This Study’s Limitations in Light of Opportunities for
Future Research on Family Medicine Faculty

This study contributes toward reducing the dearth of research studies conducted over the
last three decades on family medicine faculty. Considerable amounts of additional
research needs to be conducted that further reduces these gaps, particularly by examining
the research socialization, research environments, as well as work activities of family
medicine faculty. Future researchers should be heartened to learn that the Society of
Teachers of Family Medicine and the American Academy of Family Physicians are now
jointly working on establishing a national family medicine faculty database and mailing
list. This database should greatly reduce some of the barriers researchers face toward
studying family medicine faculty. Additionally, this database may also offer a wealth of
descriptive data that may be useful to compare with the descriptive data generated as part

of this study.

123

In addition to general studies of family medicine faculty, researchers should further
examine family medicine faculty research productivity to address some of the limitations
of this study mentioned earlier in Chapter 1. For one, as this study excluded non-clinical
family medicine department faculty, a study of this faculty group would be interesting,
particular when using elements of this study’s conceptual model. Researchers also might
want to consider using this model with other specialty’s departmental faculty to see if
elements of the conceptual framework are generalizable to pursue across specialty
boundaries. Additionally, with limited adaptations, this framework could also be used for
departmental studies in higher education. It would be particularly interesting to compare
the data gathered in higher education on the research socialization and research
environments with that found in this study to better understand that which binds, as well

as separates, medical faculty from mainstream higher education faculty.

AS this study addressed research productivity in the short-term, it would also be useful
for researchers to consider extending this study of family medicine faculty research
productivity toward a longer period of time to examine faculty career research
productivity. As family practice and family medicine now are entering their third decade
of existence, examining career productivity may now be obtainable for researchers

interested in such longitudinal studies.

Finally, as Creswell says, researching determinants of faculty research productivity

remains “puzzling” because of the large number of independent variables that are both

correlated with productivity as well as correlated with each other. This is supported by

124

the data in Appendix I of this study that shows the high degree of correlation between the
predictors of family medicine research productivity. Since this study’s research questions
focused more narrowly on predicting the research productivity of family medicine faculty
in aggregate as a discipline, without attempting to control or test for the mediating effects
of both demographic and non-demographic independent variables, follow-up studies
should be conducted using differing methodology. These studies should further refine the
model developed in this study into both the direct and mediating effects upon
productivity in light of these demographic and non-demographic interrelationships. One
blueprint provided in the literature that may be useful for researchers toward examining
direct and mediating effects was developed by Blackburn and Lawrence (1995) with

higher education faculty.

Finally, researchers need to also consider deconstructing this family medicine faculty
research productivity model by conducting more in-depth studies on individual elements
of the model, particularly in the areas of research socialization, research environments,
and psychological and cognitive characteristics of faculty. Every consideration should be
given toward employing a range of differing methodologies, especially qualitative
research methods that will provide a richer description of these areas. Only by fully
understanding how family medicine faculty are socialized and acculturated toward
research activities, can administrators and leaders in family practice and family medicine
further foster the development of researchers and the production of research needed by

the specialty, academic discipline, and society-at-large.

125

APPENDICES

126

APPENDIX A

SURVEY COVER LETTER

127

Dear Dr. <Insert Name>:

One of the many opportunities that come to the President of ADFM is the chance to work
with others who have a genuine interest in advancing the academic side of our discipline.
Several months ago, I have had the pleasure of working with Mr. Joseph Brocato, a doctoral
candidate at Michigan State University, as he developed a national study of faculty in
Departments of Family Medicine. His project has now received funding from the American
Academy of Family Physicians Foundation (AAFP/F). The purpose of this letter is to ask
for your participation.

This study will gather data on the research activities of individual family medicine faculty
and the research environments within their respective departments/divisions of family
medicine at each of the 125 medical schools and 11 geographically separated campuses in
the United States. I have reviewed the research methodology with Mr. Brocato and believe
this work will add important detail to ADFM’s recently completed department survey. The
study instrument will require only 10-15 minutes to complete. Data gathered as part of this
study will be reported in aggregate only, with no individual respondent or department being
connected to any particular responses.

The results of this research project will provide us all with a better understanding of our
discipline as a national academic enterprise, as departments within our local medical school
environments, and as individual faculty members and investigators.

Please take the few minutes required to complete the enclosed survey and return it to in the
postage-paid envelope by <insert date>. If you would like to receive a copy of the study
results, please also check the box on the back of the survey instrument.

Thank you for your time and effort.

Sincerely,

John Dickinson, MD.
President, Association of Departments of Family Medicine

Joseph J. Brocato
Associate Director of Medical Education, Riverside Osteopathic Hospital
STFM Member and Doctoral Candidate, Michigan State University

128

APPENDIX B

SURVEY INSTRUMENT

129

Family Medicine Faculty Research Survev”

 

 

*Funded by the American Academy of Family Physicians Foundation and Endorsed by
The Association of Departments of Family Medicine

 

Earl I: Prior Research Training.

Directions: Circle the o_ne_ numbered response for each question that best describes your research training
prior to yopr first faculty appointment.

 

 

Strongly Disagree Neither Agree Strongly Not
Disagree Agree! Agree Applicable
Disagme

 

I. The medical school I
attended was known for its (1) (2) (3) (4) (5) (6)
research

 

2. The institution I attended
during my residency was
well known for its research (1) (2) (3) (4) (5) (6)

 

3. The graduate school I
attended was well known (1) (2) (3) (4) (5) (6)
for its research

 

4. I received valuable
mentoring in research
related activities prior to my (1) (2) (3) (4) (5) (6)
first faculty position

 

5. I received focused research
training as part of a national
fellowship or post-doctoral
program before my first (1) (2) (3) (4) (5) (6)
faculty position

 

6. My research skills prior to
my first faculty position
included substantial training
in basic statistics, research (1) (2) (3) (4) (5) (6)
design, and data collection

 

7. I participated in meaningful
research projects before my
first faculty position (1) (2) (a) (4) (5) (6)

 

8. Journal article(s) that I
authored (or co-authored)
prior to my first faculty
position were published in (1) (2) (3) (4) (5) (6)
well-respected journals

 

9. I gave presentation(s) at
national conferences
highlighting results of
meaningful research prior to (1) (2) (3) (4) (5) (6)
my first faculty position

 

 

 

 

 

 

 

 

 

130

Part II: Current Research Interests & Activities.

Directions: Circle the % numbered response for each question that best describes your current

research interests and activities.

 

Strongly
Disagree

Disagree

Neither
Agree!
Disagree

Agree

Strongly
Agree

Not
Applicable

 

10.

Conducting research is
personally satisfying to
me

(1)

(2)

(3)

(4)

(5)

(6)

 

11.

Because of my multiple
faculty roles (including
clinical roles). I feel
pressure to find time to
engage in research

(1)

(2i

(3)

(4)

(5)

(5)

 

12.

I would describe myself
as having a well-defined
research agenda/plan

(1)

(2)

(3)

(4)

(5)

(6)

 

13.

I understand the
research guidelines for
promotion and/or tenure
within my academic
department

(1)

(2)

(3)

(4)

(5)

(5)

 

14.

| stay 'up-to-date' on
the current literature in
my research interest
areaIs)

(1)

(2)

(3)

(4)

(5)

(5)

 

15.

Since accepting my first
faculty position, I have
further developed my
research skills

(1)

(2)

(3)

(4)

(5)

(6)

 

 

16.

l have multiple research
projects under way that
are suitable for
publication

 

(1)

 

(2)

 

(3)

 

(4)

 

(5)

 

(5)

 

Part III:

urrent Res rch Environments

Directions: Circle the gig numbered response for each question that best describes your current
environments for research.

 

Strongly
Disagree

Disagree

Neither
Agree]
Disagree

Agree

Strongly
Agree

Not
Applicable

 

17.

l have a well developed
network of faculty
colleagues outside my
academic department in
family medicine with whom
I frequently discuss
research

(1)

(2)

(3)

(4)

(5)

(5)

 

 

18.

I believe that family
medicine as a discipline
places a substantial
emphasis Upon research

 

(1)

 

(2)

 

(3)

 

(4)

 

(5)

 

(5)

 

131

 

 

 

Strongly Disagree Neither Agree Strongly Not
Disagree Agree] Agree Applicable
Disagree

 

19. I believe that my academic
institution is well known
for its research (1) (2) (3) (4) (5) (6)

 

20. I believe that my academic
institution places a
substantial emphasis upon (1) (2) (3) (4) (5) (6)
research

 

21. I believe that my academic
department is well known
for its research (1) (2) (3) (4) (5) (6)

 

22. I believe that my academic
department places a
substantial emphasis upon (1) (2) (3) (4) (5) (6)
research

 

23. My academic department's
chairperson is very
supportive of my research (1) (2) (3) (4) (5) (6)
eflons

 

24. I have adequate resources
within my academic
department (such as
secretarial support,
research assistants, (1) (2) (3) (4) (5) (6)
computers, library
materials, etc.) to conduct
research

 

25. My academic department
provides me with
adequate protected time to (1) (2) (3) (4) (5) (6)
conduct research

 

 

26. A large portion of my
academic department's
faculty can be considered (1) (2) (3) (4) (5) (6)
'productive researchers'

 

 

 

 

 

 

 

Part IV: R arch P tiv

Directions: For the following section, please answer each question by circling the one number that describes your
research productivity within the following research areas.

 

 

27. Over the last two academic years, how many peer- 0 1 2 3 4 25
reviewed manuscripts have you submitted for
potential journal publication?

 

28. Over the last two academic years, how many peer- 0 1 2 3 4 25
reviewed manuscripts have you had accepted for
joumggublication?

 

29. Over the last two academic years, how many peer- 0 1 2 3 4 25
reviewed proposals/papers have you submitted
that were accepted for national conference
presentations that involve your research?

 

30. Over the last two academic years. how many 0 1 2 3 4 25
national government grant proposals have you
submitted in which you were an investigator?

 

 

 

 

132

 

31. Over the last two academic years, how many 0 1 2 3 4 25
national government grant proposals have you
submitted in which you were an investigator that
were partially or fully funded?

 

32. Over the last two academic years, how many 0 1 2 3 4 25
national private grant proposals (i.e., from
private foundations, etc.) have you submitted in
which you were an investigatofl

 

33. Over the last two academic years, how many 0 1 2 3 4 >5
national private grant proposals (i.e., from
private foundations, etc.) have you submitted in
which you were an investigator that were partially
or fully funded?

 

 

 

 

Part V: Demographics

34. What advanced degrees do you hold? (Check a_H that apply): llM.D. [10.0. DOther health doctorates
uPh.DJEd.D {Masters degree in health-related areas [Masters degree in education-related areas llOther
Masters degree

35. What is your primary medical specialty as identified by the American Board of Medical Specialties (ABMS)?
U Family Practice llOther: [Not board certified in any medical specialty

36. In what year were you born?
37. What is yourgendefl 01 Male Ch Female

38. What is your ethnic origin? (Check one)

Ch American Indian or Alaskan Native
Asian or Pacific Islander

Black, not of Hispanic Origin

Mexican American or Chicano (Hispanic)
Puerto Rican (Hispanic)

Other Hispanic

White, not of Hispanic Origin

 

999999

39. In what year were you first hired into a faculty position?

 

40. What is your current academic status? (Check one)

 

C11 Tenured

D: On the tenure track

Cb Not in a tenure track position
D4 Other

41. What is your current academic rank? (Check one)
01 Instructor

Ch Assistant Professor

Ch Associate Professor

D4 Professor

Us Clinical Instructor

Us Clinical Assistant Professor
C17 Clinical Associate Professor
Cl: Clinical Professor

Us Other (specify)

 

42. How many hours per week on average do you actually spend on research activities (such as designing research
projects, collecting research data, preparing for research conference presentations, writing journal articles, meeting
with colleagues on research projects, etc.)? Hours Per Week

43. I would like to receive the results of this study at its conclusion: Yes CI No El

133

APPENDIX C

INDEPENDENT VARIABLES: FREQUENCY DISTRIBUTIONS

134

Table 5.1

Independent Vambles: Frequency Distributions

Psychological and Cognitive Characteristics

 

 

 

 

 

 

 

 

Psychological Survey Question quency of Resume
, , Strongly Disagree Neither Agree Strongly
and Cognitive Disagree Agree/ Agree
Characteristic D'sag’ee
Motivation to Conducting research is 5.6% 18.8% 21.0% 31.7% 23.0%
do research personally satisfying to (25) (84) (94) (142) (103)
me.
Degree of role Because of my multiple 2.7% 9.8% 7.3% 30.2% 50.0%
stress faculty roles (including (12) (44) (33) (136) (225)
clinical roles), I feel
pressure to find time to
engage in research.
Research I would describe myself as 21.7% 37.4% 16.8% 15.3% 8.8%
agenda havmg a well-defined (98) (169) (76) (69) (40)
research agenda/plan.
Clear I understand the research 3.1% 12.1% 16.8% 41.5% 26.5%
expectations guidelines for promotion (13) (51) (7]) (175) (I '2)
for promotion and/or tenure within my
and/or tenure academic department.

 

135

 

 

Psychological

Survey Question

quengy of Rasmnse
Strongly Disagree Neither Agree Strongly

 

and Cognitive Disagree Agree/ Agree
Characteristic Disagree
In-depth I stay “up-to-date” on the 1.7% 11.0% 21.3% 46.9% 19.1%
knowledge of current literature in my

research area

research area(s).

(7) (45) (87) (192) (78)

 

 

 

of specialty
Research Since accepting my first 9.2% 16.9% 11.0% 37.5% 25.4%
training faculty posrtion, I have (42) (77) (50) (171) (116)
further developed my
research skills.
Multiple I have multiple research 29.6% 27.3% 10.7% 20.7% 1 1.6%
research projects under way that are (130) (120) (47) (91) (51)
projects suitable for publication.

 

 

 

136

 

Cumulative Advantage (Prior Research Socialization)

 

 

 

 

 

 

 

 

Cumulative Survey Question Fm uency of Respgnse
Strongly Disagree Neither Agree Strongly
Advantage Disagree Agree/ Agree
Factors Disagree
Prestige of The medical school I 6.4% 20.0% 20.4% 25.7% 27.4%
Medical School attended was known for its (30) (94) (96) (121) (129)
research.
Prestige of The institution 1 attended 18.9% 30.0% 20.4% 16.5% 14.1%
Resrdency during my resrdency was (87) (138) (94) (76) (65)
Program well known for its
research.
Prestige of The graduate school I 5.4% 13.1% 19.0% 22.6% 39.9%
Graduate attended was well known (9) (22) (32) (38) (67)
School for its research.
Mentorship for I received valuable 34.4% 30.9% 9.2% 16.4% 9.0%
Research mentoring in research (157) (141) (42) (75) (41)
related activities prior to
my first faculty position.

 

137

 

 

Cumulative

Survey Question

quengy of Reemnse

Strongly Disagree Neither Agree Strongly

 

 

 

 

Advantage Disagree Agree/ Agree
Factors Disagree
Fellowship/Post I received focused 48.6% 27.7% 2.8% 8.7% 12.2%
-Doctoral research training as part of
(191) (109) (11) (34) (48)
Training (with a national fellowship or
Research Core) post-doctoral program
before my first faculty
position.
Prior Research My research skills prior to 42.1% 31.9% 7.6% 9.4% 9.0%
Skills my first faculty posrtion (193) (146) (3 5) (43) (41)
included substantial
training in basic statistics,
research design, and data
collection.
Prior Research I participated in 33.2% 26.3% 10.0% 19.0% 11.5%
Project meaningful research (150) (119) (45) (86) (52)
Experience projects before my first
faculty position

 

 

 

138

 

 

Cumulative

Survey Question

quency of Respgnse
Strongly Disagree Neither Agree Strongly

 

 

 

 

Advantage Disagree Agree/ Agree
Factors Disagree
Prior Research Journal article(s) that I 41.4% 22.1% 6.5% 19.0% 1 1.0%
Productwrty authored (or co-authored) (146) (78) (23) (67) (39)
prior to my first faculty
position were published in
well-respected journals
Prior Research I gave presentation(s) at 53.1% 23.7% 4.1% 10.6% 8.5%
Productivity national conferences (206) (92) (1 6) (41) (33)

 

highlighting results of
meaningful research prior

to my first faculty position

 

 

139

 

Research Environments: Disciplinary Environment for Research

 

 

 

 

 

 

Disciplinary Survey Question quency of Resmnse
. Strongly Disagree Neither Agree Strongly
Envrronment Disagree Agree/ Agree
Factors Disagree
Professional 1 have a well developed 29.1% 36.8% 12.1% 14.6% 7.3%
Networks network of faculty (127) (161) (53) (64) (32)
colleagues outside my
academic department in
family medicine with
whom I frequently discuss
research
Disciplinary I believe that family 14.1% 49.1% 19.9% 15.4% 1.5%
Norms and medicine as a discipline (65) (227) (92) (7]) (7)
Expectations places a substantial
for Research emphasis upon research.

 

140

 

Research Environments: Institutional Environment for Research

 

 

 

 

 

 

Institutional Survey Question Freguency of Response
. Strongly Disagree Neither Agree Strongly
Envrronment Disagree Agree/ Agree
Factors Disagree
Prestidge of I believe that my academic 8.5% 19.2% 18.5% 31.6% 22.2%
Institution institution 15 well known (39) (88) (85) (145) (102)
for its research.

Institutional I believe that my academic 5.5% 13.1% 15.3% 39.3% 26.9%
Norms and Institution places a (25) (60) (70) (180) (123)
Expectations substantial emphasis on
for Research research.

 

141

 

Research En vironmenm: Departmental Environment for Research

 

 

 

 

 

 

Departmental Survey Question FLeguengy of Resmnse
. Strongly Disagree Neither Agree Strongly
Envrronment Disagree Agree/ Agree
Factors Disagree
Prestidge of I believe that my academic 14.4% 33.4% 22.7% 21.2% 8.3%
Department department IS well known (66) (153) (104) (97) (38)
for research.
Departmental I believe that my academic 8.2% 22.7% 27.8% 32.0% 9.3%
Norms and department places a (37) (103) (12 6) (145) (42)
Expectations substantial emphasis upon
for Research research.
Mentorship My academic 4.2% 15.1% 25.7% 32.3% 22.8%
and Leadership department 5 chairperson (16) (57) (97) (122) (86)
for Research is very supportive of my
research efforts.
Research I have adequate resources 18.0% 23.1% 18.9% 29.6% 10.4%
Support within my academic
(78) (100) (82) (128) (45)
(Resources) department (such as

 

secretarial support,
research assistants,
computers, library
materials, etc.) to conduct

research.

 

 

142

 

 

Departmental

Survey Question

quency of Respgnse
Strongly Disagree Neither Agree Strongly

 

 

 

Environment Disagree Agree/ Agree
Factors Disagree
Protected Time My academic department 32.8% 35.4% 16.2% 10.2% 5.5%
for Research provrdes me With adequate (138) (149) (68) (43) (23)
protected time to conduct
research.
Productive A large portion of my 25.8% 42.8% 18.8% 10.8% 1.8%
Colleagues academic department 5 (115) (191) (8 4) ( 48) (8)

 

faculty can be considered

“productive researchers.”

 

 

143

 

APPENDIX D

DEPENDENT VARIABLES: FREQUENCY DISTRIBUTIONS

144

Table 5.2
Dependent Variables: Frequency Distributions
Manuscripts Submitted for Publication

requency

Submitted

 

Sor

Manuscripts Accepted for Publication

requency Percent
Submitted Percent

Sor

 

Proposals/Papers Accepted for Conference Presentations

requency ercent
Submitted Percent

 

145

National Government Grant Proposals Submitted

requency ercent

Submitted

01‘

National Government Grant Proposals Funded
Frequency ercent
Submitted

5
1

National Private Grant Proposals Submitted

requency
Submitted

1

146

 

 

Percent

 

National Private Grant Proposals Funded

requency ercent
Submitted Percent

 

147

APPENDIX E

MEAN PRODUCTIVITY DIFFERENCES BY AGE

148

Table 5.3

Mean Productivity Differences bsLAgg

Tukey HSD
(I)
AGEGRP
Manuscripts Accepted for
Publication 30-39
40-45
46-50
51+
Proposals/Papers Accepted for
Conference Presentations 30-39
40-45
46-50
51+
National Government Grant
Proposals Funded 30-39
40-45
46-50

149

(J)

AGEGRP

40-45
46-50
51+
3-39
46-50
51+
30-39
40-45
51+
30-39
40-45
46-50

40-45
46-50

51+
30-39
46-50

51+
30—39
40-45

51+
30-39
40-45
46-50

40—45
46-50
51+
30—39
46-50
51+
30-39
40-45
51+

(I-J)

-.85*
-.90"‘
-.76*
.85*
.05
.09
90*
.05
.14
.76*
-.O9
-.14

-.60*
-.35
-.30
.60*

.24
.30
.35
-.25
. 05
.30
-.30
-.05

-.27
-.35*
-.33*

.27

-.08

-.06

.35“

.08
.02

.18
.20
.19
.18
.19
.18
.20
.19
.19
.19
.18
.19

.17
.18
.17
.17
.17
.16
.18
.17
.17
.17
.16
.17

.11
.11
.11
.11
.11
.11
.11
.11
.11

Mean Diff. Std. Error Sig.

00

.00
.00
.00
.99
.96
.00
.99
.88
.00
.96
.88

.00
.20
.30
.00
.47
.26
.20
.47
.99
.30
.26
.99

.06
.01
.01
.06
.89
.93
.01
.89
.99

(I)
AGEGRP
5 1+

National Private Grant Proposals

Funded 30-39

40-45

46-50

51+

(J)

AGEGRP

30-39
40-45
46-50

40-45
46-50

51+
30-39
46-50

51+
30-39
40-45

51+
30-39
40-45
46-50

* The mean difference is significant at the .05 level.

150

(1-1)

.33*
.06
-.02

-.20*
-.1 l
-.11
.20*

.09
.08
.11
-.O9
-.01
.11
-.08
.01

.11
.10
.11

.07
.08
.08
.08
.08
.08
.08
.08
.08
.08
.08
.08

Mean Diff. Std. Error Sig.

.02
.93
.99

.04
.52
.46
.04
.67
.66
.52
.67
1.00
.46
.66
1.00

’—

APPENDIX F

MEAN PRODUCTIVITY BY ADVANCED ACADEMIC DEGREE

151

Table 5.4

Mean Productivity by Advanced Academic Degree

Manuscripts Proposals/Papers National
Accepted for Accepted for Government
Publication Conference Grant

Presentations Proposals

Funded
MD only Mean .85 .55 .354
N 309 309 308
Std. Deviation 1.39 1.1 19 .759
D0 only Mean .51 .60 .10
N 47 47 47
Std. Deviation .88 1 .21 .3 1
MD or DO Mean 1.48 1.29 .68
Plus Other N 96 97 97
Degrees Std. Deviation 1 .66 1 .70 1 .09

152

National
Private
Grant
Proposals
Funded
.20
307
.58
.06
47
.25
.38
94

.67

APPENDIX G

MEAN PRODUCTIVITY BY ACADEMIC RANK

153

Table 5.5
flan Productivity by Academic Ran—k
Manuscripts Proposals/Papers National

Accepted for Accepted for Government
Publication Conference Grant

Presentations Proposals
Funded

Assistant
Professor Mean .57 .61 .21
N 162 162 161
Std. Deviation 1 .05 1 .22 .55

Associate
Professor Mean 1 .53 1 .08 .62
N 89 9O 90
Std. Deviation 1.76 1.50 1 .01
Professor Mean 2.04 1 .34 .87
N 70 70 70
Std. Deviation 1 .74 1.69 1 .19

Clinical

Academic Mean .48 .26 .21
Ranks N 120 120 120
Std. Deviation .91 .69 .54

154

National
Private
Grant
Proposals
Funded

.15
160
.44

.31
9O
.61

.47
68
.84

.11
119
.39

APPENDIX H

MEAN PRODUCTIVITY BY YEARS OF EXPERIENCE

155

Table 5.6

Mean Productivity by Years of Experience

Years as
faculty
member

1-5 Years

6-10 Years

11-15 Years

16 or More
Years

Mean
N
Std. Deviation

Mean
N
Std. Deviation

Mean
N

Std. Deviation

Mean
N
Std. Deviation

Manuscripts Proposals/Papers

Accepted for
Publication

.35
124
.77
.83
122
1.35
1.38

80
1.84

1.40
113
1.48

156

Accepted for
Conference
Presentations

.38
124
.89

.64
122
1.25

1.16
81
1.63

.88
113
1.40

National
Government
Grant
Proposals

Funded

.15
124
.44

.29
121
.66

.57
81
1.01

.67
113
1.05

National
Private
Grant
Proposals
Funded

.073
123
.32

.28
120
.65

.39
81
.75

.22
111
.55

APPENDIX I

PEARSON CORRELATIONS FOR REGRESSION MODEL VARIABLES

157

693$ .22 mod 05 a 285:9» a 856:8 ..
623.3 .96. So as a 285:9... a 530:8 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8338> 332 common»; .5.“ mucus—2.50 :8..qu
Nam 033.

3* En m9. 3» 8m 2
. So. So. So. So. 625.8 an $68.33
83 :68. 8... ram. £8. 832.8 Samoa .2 Sao8< announces.
Ba No... an 8m 8m 2
8o. . So. 8o. «8. 625$ em
:63. 83 :3”. :EN .3. 83230 828a 688m .2 ”0283”.
3. Sn :4 8” Sn 2
So. So. o8. «8. 625.3 .65 55535
80. :20 83 .3. :22. 5.3.2.8 Samoa 568mm 52.5
BM 8” on» 8m 8m 2
8c. So. So. . o8. 62.9.3 .65 «>558
:muo. £8. .3. 83 2.8”. 58.88 828a 2m 3866?.
8m 8m Sm 8m in 2
ace. «8. «8. 8o. . A393 .5
L8. .3. 2.8.. imam. 83 caisson Samoa uses: .25
cos—3.35 neaomom .8 E95835 mica ImEEEP Eta J
.3 3308.4. 39331 52331 fie.mo.oco>mn_
automscmz €2.50
Eons—goo

158

625.3 .82 8.: 05 .0 0:85:05 2 802080 ..
62.5.3 .92 3.: 05 5.. 0:85:05 2 8:22:00 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we «mm. 0?. 08. non. 2 00058000000
8: o: 8: 08 625.3 .20 8:20.50 :2 8.082
08.5 {09. Re. Imam. tuna. 5050.05.60 002000 800003000020
own «on Em own Bu 2
n8. . 8:. 8:. «8. 625.3 .20
:09. coo; tnnn. :EN. .09. 550.0500 002000 50.00001 :8 00050001
0? Sn 36 000 Sn 2
2 F. :8. 8:. 8:. 2852.3 .20 0:05.555
2.0. town. coo; .mm F. You 5. 550.0500 002000 5050000”. 5.2.50
can own «an new cow 2
:8. 8:. :8. . :8. 625.3 .20 25:80
20%. :EN. .09. 084 imam. 0050.230 002000 0cm .00_mo_o.._0>00
Sm 000 Sn mom in z
08. «8. «8. :8. . 625.3 .20
:03. .8 w. tom ... imam. coo; 00:0.0tow 003000 0550:. 00:0
«0058:0020 220003. :2 0008:0035 025030 050 [9558... .050
0000:0200 :2 0023300 00.00000 30.00.2030
00000090 20 0:05.50
0002000080

 

 

 

2.000.230

159

00.3.0. .05. 00.0 05 .0 0:000:90 0. 800.0000 ..

60.5.0. .90. 3.0 0:. .0 08000.0 0. 800.0000 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000 000 000 000 000 z
. .00. N00. 000. 000. 82.8.0. 0.0 000:0“. 0.000023 00.0
000.. :00 w. 30. :03. :00 0. 00.00.0000 000.003 «00.000030 300002
000 N00 30 0mm com 2
.00. . 000. 000. ~00. 62.3.0. 0.0
:03. 000.0 :05. :EN. .00.. 02.0.0000 000003 0000001 .8 000000000.
000 000 03 000 30 2
~00. 000. 000. N00. 62.5.0. 0.0 0055.20
3.0. :05. 000.. .09. 1000. 00.00.0000 000.003 00.00000. E0000
000 0% 000 wow wow 2
000. 000. 0N0. . 000. 62.9.0 0.0 02.0000
:03”. :EN. 03.. 000.? 2.000. 02.0.0000 000.003 000 00.00.0503
000 00m 000 00.0. 30 z
000. N00. ~00. 000. . 89.073 0.0
mm... .00 F. :00 P. .5000. 000.. 00.00.0000 000.003 00.0.0.0. .003
000003 0000000”. .8 .00800..>0w 02..0m0|000|0 40.0.0; .003 I
0.0000003 0005000”. 00000000. 30.00.002.03
.0000 000.000
.00EE0>00
.00802

 

 

000.00.00.00

160

62.2.0 .22 00.0 0:. .0 002.000 0. 8020000 ..
62.2.8 .22 .00 0:. .0 0005:...» 0. 8022.00 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.00 000 ~00 000 000 z
. 000. 000. 000. 000. 62.2.0 0.0 000:... 200000...
000.. :0 .~. 2.00.. :00... 100.. 02.0.0000 000.003 .020 0.02.3 .3202
000 ~00 .00 0~n 00~ z
000. . 000. 000. ~00. 62.0.3 0.0
:m.~. 000.. 2.000. $.0~. .00.. 02.0.0000 000.003 00.00001 .0. 00050001
~00 .00 .3. 000 .00 z
000. 000. 0~0. ~00. .00..0..~. 0.0 .00200..>0m
:09. .5000. 000.. .0~ .. 100.. 02.0.0000 000.003 00.00001 .0250
000 0~0. 000 000 00~ z
000. 000. 0~0. . 000. .00..0..~. 0.0 0>...0000
:09. 3.0m. .m~.. 000.. .3000. 02.0.0000 000.003 000 30.00.002.03
.00 00~ .00 00~ 0.0 2
000. ~00. ~00. 000. . 62.3.0 0.0
:00 .. .00 .. :00 .. {000. 000.. 02.0.000w 000.003 00.0.0. .0 .003
0000: 3 00.00001 .0. 00.002500 0%[0220F .003
0.000023 00050001 00.00001 320200003
.020 0.0>..3 .00....0
.000..0z

 

 

000322.30

161

..00_.0..~. .05. .00 0... .0 0000.00.» 0. 02.0.0000 .:
..00_.0..~. .20. 00.0 0... .0 0000.000 0. 800.0000 ..

 

 

 

 

 

 

 

 

 

000 ~00 000 z
000. .00. €2.50. 0.0 02.02303 .0.
000.. .00.- $00 .. 02.0.0000 000.003 00.308... 0.3000000...
~00 ~00 .00 z
000. . 000. .00..0..~. 0.0 .2000. 00.0005 00
.00.- 000.. .~0.. 02.0.0000 000.003 00.00001 00.2 0.0.03
000 .00 000 z
.00. 000. . .00..0..~. 0.0 00.00000
:00 .. .~0 .. 000.. 02.0.0000 000.003 020.000. 00..00.3I
02.02.33 SE10. 00.00001
.0. 00.3002 00.3005 00.2000.
0.3000000: 00 00.00001 00000.3
00.2 0.0.0.

 

 

00000-0000

162

625$ .26. 3d 3 a 585:9... a. 8.3.2.8 .:

.ao__s-~v .92 3.. a5 a 285ch a 832.8 ..

 

 

 

 

 

 

 

 

 

 

 

 

 

3v om... Ev o3. 03 z

. 80. 8o. QB. o9. 623$ gm 82E 2889a

coo; zoom. :58. 8o: woo. 5:59.00 523.... E80 Suzi .mcosmz

omv m9. mm». «9. v3. 2

8o. . 8o. 80. «up. A8__s.~v gm BEE 2389a 580
$09.... oooé 12m. 192.. who. coaflotoo 323a EoEEo>00 .mcozmz

Ev 3:. 8v «9. m3

2

a... 33.5 scésfifim
:mwm. :mnm. coo; .3 Fr :63. 5.6.9.30 comfioa 3383.389»;

93. Nov «9. «cc Ev 2

Sb. So. as. . o8. €2§$ gm :2me. 2:485

one: 1.8...- .2 Fr coo.“ .Nop. cozmfitoo Samoa as 38331 was. 2:5“.

o3 3% 9% Se 2:. z

2:. .‘fi. 8o. o8. . 825.3 gm 6.8mm”.

«8. «B. :63. .we. r oooé 9.26.2.6 :3..me 9.289. mmzmmtl

conga“. vanes“. 383535 65mg cemomom
28805 $338.... 3:23:00 .2 $5335 952qu
E80 995 EEO umfimoo< 20 mm cemwmom ooamen.
.20st 5055900 awn—2332a 35. ESE
.mcozmz

 

 

2335230

163

28.3.3 .22 8.0 05 a 285:2» 2 5:22.00 ..
623$ .92 8.0 as .0 285:2» 2 8:22.00

 

 

 

 

 

 

 

 

 

 

 

 

$4 mmm son 2
0~0. ~00. 62.3.8 .20 .coEcESCm
oooé .mNF. toot 00:50:00 000.000 5.0000”. «02.50
Sn wow mom 2
owe. . 08. 623.8 .20 95:08
89. oooé 10mm. 00:30:00 000.000 0:0 3030.053.“
Sm com in z
«8. 8o. . 625$ gm
tow _.. (mam. coo; 00:90:00 000800 95:8... Btu
EoEcQScm 0>£cao 0:0 Imc_c_m._._. 00:0
20.00001 _3_oo_0;0>mn_
.0250

 

 

30:20:00

164

625$ .92 00.0 05 50 0:85:90 0_ 8:20:00 ..
625$ .32 3.: 05 .0 0:85:90 2 8:20:00 .:

 

 

 

 

 

 

 

 

 

 

 

~00 Ev 50 z
00:. 8:. 625$ .90 6.0105 05.820
08; .we. :03. 5050.05.00 502001 00 5200001 005. 5:501
$0 000 000 z
00:. . v00. 625$ .90 .5881
.we... 08.5 50. 5050.05.00 502001 952090.. 090020.
50 000 ~00 z
08. 3.0. . 625$ .90
:03. 50. 000.? 5050.05.00 502001 5200001 :00 002000041...
SE18 5200001 5200001 :2
05.0005 052090.. 00200001
00 5200001 095020
022 55:0”.

 

 

050.00.05.50

165

.A00._0«.Nv _0>0. 3.0 05 .0 55005590 0. 5030.05.50 .2.

 

 

 

 

 

 

 

 

 

 

 

Ev 000 30 00v 2
. :8. 08. 08. 625$ .90 00200 0.80520
coo; 2.000. 2.000. 2.000. 5050.0500 502000 05050 0.0050 .050502
000 000 000 000 z
8:. 8:. :8. 625$ .90 020:0... 28820 520
:80. 000.5 :20. 2.000. 5000.0500 502000 050.55.05.00 .050502
50 00v 000. 000. z
. . . 050505500050
coo 000 000 30:05-8 .90 0050500500 :0. 00.0002
.000. :20. 000.0 :30. 50.00.0500 502000 2000075000020
000 00v 000. 000 z
:8. 80. 80. . 625$ .90 8.085000
2.000. 000. :30. 000.5 5000.050w 502000 .0. 00.0080. 205000505.
00050.... 00050... 0505050020 50502500
0.000020 0.000020 005050500 :2 :0. 00.0002
.520 0.0050 .505 00.00090 20 0005000505.
.050502 50550000 0002000020
.050002

 

 

30.00.0500

166

625$ 52 00.0 05 .0 0:85:90 0. 5.50:8 ..
625$ .22 8.: 05 .0 0:85:90 0. 8.50:8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00v 00... 03. 000 000 z
. 08. 80. 08. to. 625$ .90 8.08.300
000.. :03 .- 2.5.00. :00m. .00 P. 50050500 502000 50.. 00.00090 205000505.
00v 00v 000 «0.. 500 z
08. . 03. 08. 8:. 625$ .90
r00..- 000.0 000. 2.08.- :Nvmv 50.00.2500 502000 500500
03. 00v 000. 03 03 z
8:. 03. 000. EL. 625$ .90 8.00001
:50. 000. coo; 30. 3.0. 5000.05.00 502000 50 0.002, :00 200...
000 N00 03. N00. «3 z
08. 08. 30. . 80. 625$ .90
...o0~. 2.08.. So. coo; 2.050. 5000.05.00 502000 5055.05. 3.0000 00 200>
P00 500 03 N3 000 z
to. 8:. EL. 08. . 625$ .90
.05 F. :Nvmr 30. .3050. 0004 50.00.0500 502000 0m<
50502300 500500 5200001 50520.2 000.
:0. 02582 :0 282, 00000
0000000505. :00 200... 00 200>

 

 

050.00.05.00

167

625$ 52 8.: 05 .0 0:85:90 0. 802080 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

.00.. «00. 03.. :3. 000. 2 08.5800...
00: :8 8: 0R 625$ .90 8:20.80 :2 02:08,:
8:.. 08.- :80. :83. 3:. 8.50:8 80:80 205000.882:
:00 000 000 N00 500 z
00:. . 03. 8:. :8. 625$ .90
8:.- :8.. 8:. 2.80.- r000: 8.02080 80:80 80:00
03 00v 00¢ 03. 03 z
8:. 03. 3:. 2L. 625$ .90 8.00005
:80. :8. :8.. 8:. 3:. :25080 80:80 8 2025 :00 002..
:3. 00.. 03. ~00 03. z
08. 8:. v8. . 8:. 625$ .90
:8.. 2.80.- .8. :8.. 050. 802080 80:80 8:522 5.80... 00 28>
000 :00 03. 03 00v 2
0k. :8. IL. 8:. . 625$ .90
3:. 1.000.- 3:. 2050. :8.. 8.02080 80:80 00<
0500050020 500500 5200001 5055.05. 0m<
00520500 :2 50 0.00; 5.000...
00.0000< 20 :00 200: 00 28>
0005000020

 

 

2033258

168

625$ .82 8.: 05 .0 0:85:90 0. 8.00.080 ..

.Aumzmfimv _0>0. 3.0 05 .0 «5005590 0. 50.50.0500 .2.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00v 00¢ 03 00v 300 z
. 000. ooo. ooo. 03. 80:000. .90 000500 0.000020 .520
oooé Fvor 2.000. {008 .0ao. 50.50.0500 502000 5505:2030 .050502
000 000. 00.. ~00 0.00 z
80. . 08. :8. 8:. 625$ .90
$5.- oooé moo. :momr 205.. 50.00.0500 502000 500500
03. 00v 00v 0!. 03. z
8:. 03. 08. 0k. 625$ .90 8.00001
.00... moo. oooé 5o. 3o. 52.0.0500 502000 50 0.00.5 :00 200...
00v N00 03 N00 «3. z
8:. 8:. v8. . 8:. 625$ .90
{00$ :88. «mo. oooé .3050. 50.00.0500 502000 50520.2 3.0000 00 200>
F00 00¢ 03. m3 000 z
08. 8:. 0R. 8:. . 625$ .90
.08. :08; 3:. :05. :8.. 8.02080 80:80 00<
000500 500500 5200001 50520.2 0d<
0.000020 50 0.00.5 3.0000
.520 :00 200.... 00 200>
550555.030
.050502

 

 

80.00.0500

169

.0250. _0>0_ 00.0 05 .0 58550.0 0. 500.050 ..
.3500 0.6. 8.0 05 .0 280.50.... 0. 5000.008 .:

 

 

 

 

 

 

 

 

 

 

 

 

 

F0¢ 0¢¢ .¢¢ 00¢ 0¢¢ z
. 000. 08. 000. 30. 62.9.0. 0.0 00050 0.00220
08.. 000.- :03. L F F. .00. 800.050 80.000 .520 0.02.0 08002
0¢¢ 00¢ 00¢ 00¢ 00¢ z
000. . 0¢0. 08. 80. 60.3-0 0.0
000.- 08.. 08. :80; .00.. 800.050 80.000 .0050
.¢¢ 00¢ 00¢ 0¢¢ 0¢¢ z
80. 0.0. ¢00. £0. 62.3.00 0.0 5.00000
1N3. moo. coo._. «mo. 30. coaflctoo :0»..an :0 x82. boa 00:01
00¢ 00¢ 0¢¢ 00¢ 0¢¢ z
000. 80. ¢00. . 80. 82.5.0 0.0
L . F. :000.- .00. 80.. 2.000. 500.050 50.000 .8222 0.30“. 00 0.09
0¢¢ 00¢ 0¢¢ 0¢¢ 00¢ z
30. 08. £0. 08. . 62.3.0. 0.0
.00. .00.. :0. 2.000. 08.. 800.230 53.000 $0
0005:... 000500 5200000 50020.... 0m<
0.000020 50 0.00.5 2.30...
050.0 000>00 .00 0501 00 200>
.050002

 

 

30.00.2000

170

00000.0. 0.6. .00 0... .0 28500.0 0. 02.0.2.8 .2.
02.0.0. .03. 00.0 o... .0 £80.00... 0. 800.230 ..

 

 

 

 

 

 

 

 

 

 

 

 

 

.00.. 00¢. 03.. 00¢. 00¢. . 2 0500850020
000 000 0.0 000 62.0.0. 0.0 8020.500 .0. 00.082

89. 1.000. 2.000. .0. .. 00o. 52.0.0200 500.000 200002000020

00.. 000 03 000. 00.. z

000. . 000. .3. 000. 62.0.0. 0.0 0008.300
:30. So. 2.000. .00o. .0o.. 52.0.0000 500.000 .2 00.008... 0.050050...

03. 03 N00 00.. 00¢ z

000. 000. 000. 0.0. 62.0.0. 0.0 .8080.
2.000. :00... ooo.. .00.. .0. .. 52.0.0200 500.000 ov0. 000.. 0.050.

000 000 000 00.. 00... z

0.0. .3. 000. . 00.. 62.0.0. 0.0 0003.00.80.08
.0. .. .000. .03. 08.. {or 50..0.0..00 500.000 00050>0<

00.. 000 000 000 00.. z

000. 000. 0.0. 00.. . 62.0.0. 0.0

00o. .09. .0. .. 2b.- ooo.. 52.0.0200 500.000 02.002000 00050>0<

0500850020 50002300 .00080. 9.0. 0.0000... 0.2002000

3520.500 .2 .0. 00.00090 .80.. 0.050. 0.000.000 00050>0<

00.00090 0.0 0.0000050... 00050>0<

000.0.000020

 

 

0.00.00.00.00

171

00.0.0. .05. .00 0... .0 £000.00... 0. 800.0000 .2.

..00_.0..~. .05. 00.0 0... .0 .0808... 0. 00.0.0000 ..

 

 

 

 

 

 

 

 

 

 

 

 

 

.0.. 00v .3. 00.. 000 z

. 000. 000. 00.. 03. 80.5-0. 0.0 0005: 0 0.00000. 0

000.. £000. 2.000. 03. .000. 02.0.0000 500.000 .505 0.02.0 .2202

000 000 03. 30 3.. z

ooo. . ooo. Noo. moo. 30:0..Nv .o_w 009.30 0.000090 30.0
:000. 000.. 2.000. 000. 2.00.. 52.0.0000 500.000 .c05.5.0>00 .050..0z

.40 0.4 00v 00v 000 z

000. 000. 000. 0.0. 80:070. 0.0 600000.
2.000. 2.000. 000.. .00.. .0... 52.0.0000 500.000 9.0. .80.. 0.050.

00.. Ev 00.. 00v 00v 2

00.. 000. 000. . 8.. 62.0.0. 0.0 000.000.000.000

03. 000. .00.. 000.. .00.- 52.0.0000 500.000 00050>0<

00.. 00.. 00.. 00v 00v 2

03. 000. 0.0. 5. . 62.0.0. 0.0
.000. 2.00.. .0... .00.- 80.. 52.0.0.3 500.000 02.002000 00050>0<

000500 000500 600000. 0v... 0.09.0... 0.20%
0.000020 0.000090 .00.. 0.350. 0.000.000 00050>0<
.505 0.02.0 .505 00050>0<
.050..0z 5085.030
.050..02

 

 

 

05000—230

172

REFERENCES

173

REFERENCES

Aday, L. A. (1989). Designing and conducting health surveys: a comprehensive guide
( lst ed.). San Francisco: Jossey-Bass Publishers.

Alasuutari, P. (1995). Researching Culture: Qualitative methods and cultural studies.
Thousand Oaks, CA.: Sage Publications.

Alguire, P. C., Anderson, W. A., Albrecht, R. R., & Poland, G. A. (1996). Resident
research in internal medicine training programs. Annuals of Internal Medicine,
124(3), 321-328.

Allison, P. D., & Stewart, J. A. (1974). Productivity differences among scientists:
evidence for accumulative advantage. American Sociological Review, 39, 596-

606.

Anderson, W. A., Stritter, F. T., Mygdal, W. K., Amdt, J. B., & Reid, A. (1997).
Outcomes of three part-time faculty development fellowship programs. Family
Medicine, 29(3), 204-208.

Applegate, W. B., & Williams, M. E. (1990). Career development in academic medicine.
American Journal of Medicine, 88(3), 263-267.

Aren, L., & Ben-David, J. (1968). Socialization and career patterns as determinants of
productivity of medical researchers. Journal of Health Social Behavior, 9(1), 3-
15.

The Association of American Medical Colleges. (1999). Faculty Roster .Washington,
DC.

The Association of Departments of Family Medicine Task Force on Clinical Practice in
US Family Medicine Departments in Academic Medical Centers. Clinical practice
in academic medical center departments of family medicine. (1997). Family
Medicine, 29(1), 15-20.

Austin, A. E. (1990a). Faculty cultures, faculty values. In W. G. Tierney (Ed.), Assessing
academic climates and cultures (pp. 61-74). San Francisco, CA.: Jossey-Bass,
Publishers.

Austin, A. E. (1990b). Tension, stress, and the tapestry of faculty lives.
Academe(January-February 1990), 38-42.

Austin, A. E. (1992). Faculty cultures. In A. I. Morey (Ed.), Encyclopedia of Higher
Education (pp. 1614-1622). Oxford, NY.: Pergamon Press.

174

Austin, A. E. (1994). Understanding and Assessing Faculty Cultures and Climates.
Unpublished Manuscript.

Austin, A. E., & Gamson, Z. F. (1983). Academic workplace: new demands, heightened
tensions (Vol. 10). Washington, DC: Association for the Study of Higher
Education.

Babbie, E. (1990). Survey research methods ( 2nd ed.). Belmont, CA: Wadsworth
Publishing Company.

Baldwin, C. D., Levine, H. G., & McCormick, D. P. (1995). Meeting the faculty
development needs of generalist physicians in academia. Academic Medicine,
70(1 Supplement), S97-103.

Bean, J. (1982). A causal model of faculty research productivity. Paper presented at the
annual meeting of the American Education Research Association, New York, NY.

Becher, T. (1987). The disciplinary shaping of the profession, The Academic Profession
(pp. 271-303). Berkeley, CA.: University of California Press.

Behmeyer, C. E. (1974). Institutional and personal correlates of faculty productivity.
Unpublished doctoral dissertation, The University of Michigan, Ann Arbor, MI.

Behmeyer, C. B., & Blackburn, R. T. (1975). Environmental and personal attributes
related to faculty productivity. Ann Arbor, MI.: Center for the Study of Higher
Education, The University of Michigan. (ERIC Document Reproduction Service
No. ED 104317)

Bentley, R., & Blackburn, R. (1990). Changes in academic research performance over
time: a study of institutional accumulative advantage. Research in Higher
Education, 31(4), 327-353.

Bess, J. L. (1978). Anticipatory socialization of graduate students. Research in Higher
Education, 8(4), 289-317.

Bickel, J. (1991). The changing faces of promotion and tenure at US. medical dchools.
Academic Medicine, 66(5), 249-256.

Biglan, A. (1973). The characteristics of subject matter in different academic areas.
Journal of Applied Psychology, 5 7(3), 195-203.

Blackburn, R. T., Behmeyer, C. B., & Hall, D. E. (1978). Research note: Correlates of
faculty publications. Sociology of Education, 51 (April), 132-141.

175

Blackburn, R. T., & Bentley, R. J. (1993). Faculty research productivity: Some
moderators of associated stressors. Research in Higher Education, 34(6), 725-
745.

Blackburn, R. T., & Fox, T. G. (1976). The socialization of medical school faculty.
Journal of Medical Education, 51 (10), 806-817.

Blackburn, R. T., & Fox, T. G. (1983). Physicians' values and their career stages. Journal
of Vocational Behavior, 22, 159-173.

Blackburn, R. T., & Lawrence, J. H. (1995). Faculty at Work: Motivation, Expectation,
Satisfaction. Baltimore, MD.: The Johns Hopkins University Press.

Blackburn, R. T., Wenzel, S., & Bieber, J. P. (1994). Minority vs. majority faculty
publication performance: A research note. The Review of Higher Education, 1 7,
271-282.

Bland, C.J., & Ruffin, M. (1992). Characteristics of a productive research environment:
Literature review. Academic Medicine, 6 7(June), 385-397.

Bland, C. J ., Hitchcock, M. A., Anderson, W. A., & Stritter, F. T. (1986). A review of
family medicine faculty development fellowship programs, their alumni and

recommendations for future programs. Proceedings of the Annual Conference on
Research in Medical Education, 25, 157-162.

Bland, C. J ., Hitchcock, M. A., Anderson, W. A., & Stritter, F. T. (1987). Faculty
development fellowship programs in family medicine. Journal of Medical
Education, 62(8), 632-641.

Bland, C. J ., & Schmitz, C. C. (1986). Characteristics of the successful researcher and
implications for faculty development. Journal of Medical Education, 61(1), 22-
31.

Bland, C. J ., Schmitz, C. C., Stritter, F. T., Henry, R. C., & Aluise, J. J. (1990).
Successful faculty in academic medicine: Essential skills and how to acquire
them. New York: Springer Publishing Company.

Bland, C. J ., Simpson, D. B., Hekelman, F. P., & Stritter, F. T. (1997). Introducing the
dedicated issue on faculty development. Family Medicine, 29(4), 231-232.

Bland, C. J ., & Stritter, F. T. (1988). Characteristics of effective family medicine faculty
development programs. Family Medicine, 20(4), 282-288.

Blau, P. M. (1973). The Organization of Academic Work. New Brunswick, NJ .:
Transaction Publishers.

176

Block, S. D., Clark-Chiarelli, N., Peters, A. S., & Singer, J. D. (1996). Academia's chilly
climate for primary care. JAM, 276(9), 677-682.

Blondell, R. D., Norris, T. E., & Coombs, J. B. (1992). Rural health and family medicine.
Family Medicine, 45(6), 2507-2510.

Bowen, H. R., & Schuster, J. H. (1986). American professors: A national resource
imperiled. New York: Oxford University Press.

Boyer, E. L. (1990). Scholarship reconsidered: priorities of the professoriate. Princeton,
NJ: The Carneige Foundation for the Advancement of Teaching.

Braskamp, L. A., & Ory, J. C. (1994). Assessing faculty work: Enhancing individual and
institutional performance. San Francisco: Jossey-Bass Publishers.

Braxton, J. M. (1983). Departmental colleagues and individual faculty publication
activity. The Review of Higher Education, 6(2), 115-128.

Braxton, J. M., & Bayer, A. E. (1986). Assessing faculty scholarly performance. In J. W.
Creswell (Ed.), Measuring faculty research performance (Vol. 50, pp. 25-42). San
Francisco: Jossey-Bass Publishers.

Burke, D. L. (1992). Physicians in the academic marketplace. New York: Greenwood
Press.

Burke, W., Baron, R. B., Lemon, M., Losh, D., & Novack, A. (1994). Training generalist
physicians: structural elements of the curriculum. Journal of General Internal
Medicine, 9(4 Supplement 1), 823-30.

Calligaro, I. L. S., & Others, A. (1991). Pharmacy research in academic institutions.
American Journal of Pharmaceutical Education, 55(1), 40-44.

Campos Outcalt, D., & Midtling, J. E. (1993). Family medicine role models at US
medical schools. Why their relative numbers are declining. Archives of Family
Medicine, 2(8), 827-832.

Cassell, E. J. (1997). Doctoring: The nature of primary care medicine. New York:
Oxford University Press.

Centra, J. A. (1981). Research productivity and teaching effectiveness. (ERIC Document
Reproduction Service No. ED215637).

Chaffee, E. E., & Tierney, W. G. (1988). Collegiate culture and leadership strategies.

New York: American Council on Education and Macmillian Publishing
Company.

177

Clark, B. R. (1983). The higher education system: academic organization in cCross-
national perspective. Berkeley, CA: University of California Press.

Clark, B. R. (1987). The academic profession. Berkeley, CA.: University of California
Press.

Clemente, F. (1973). Early career determinants of research productivity. American
Journal of Sociology, 79(2), 409-419.

Cole, J. R., & Singer, B. (1991). A theory of limited differences: Explaining the
productivity puzzle in science. In H. Zuckerman & J. R. Cole & J. T. Bruer
(Eds.), The outer circle: Women in the scientific community (pp. 277-310). New
York: Norton.

Collins, B. (1993). A review and integration of knowledge about faculty research
productivity. Journal of Professional Nursing, 9(May-Jun), 159-168.

Colwill, J. M. (1986). Academic departments of family medicine: A retrospective and a
prospective view. Family Medicine, 18(6), 384-3 88.

Colwill, J. M. (1987). Family medicine in the medical school. In W. J. Doherty & C. E.
Christianson & M. B. Sussman (Eds), Family medicine: The maturing of a
discipline (pp. 71-85). New York: Haworth Press.

Crane, D. (1965). Scientists at major and minor universities: A study of productivity and
recognition. American Sociological Review, 30, 699-714.

Creamer, E. G. (1998). Assessing faculty publication productivity: Issues of equity ( Vol.
26). Washington, DC: The George Washington University, Graduate School of
Education and Human Development.

Creswell, J. W. (1985). Faculty research performance: Lessons from the sciences and the
social sciences. ASHE-ERIC Higher Education Report No. 4, I 985.

Creswell, J. W. (1990a). Conceptual explanations of research performance. In M. J.
Finklestein (Ed.), The ASHE Reader on Faculty and Faculty Issues (2nd ed., pp.
240-248). Needham Heights, MA.: Ginn Press.

Creswell, J. W. (1990b). Correlates of research performance. In M. J. Finklestein (Ed.),
The ASHE Reader on Faculty and Faculty Issues (2nd ed., pp. 249-262).
Needham Heights, MA.: Ginn Press.

Creswell, J. W. (1994). Research design: Qualitative and quantitative approaches.
Thousand Oaks, CA: Sage Publications.

178

Culpepper, L. (1991). Family medicine research: major needs. Family Medicine, 23(1),
10-14.

Culpepper, L. (1993). Family medicine research: need for capacity-building. Rhode
Island Medicine, 76(6), 303-308.

Culpepper, L., & Becker, L. A. (1987). Family medicine research: two decades of
developing its base. In W. J. Doherty & C. E. Christianson & M. B. Sussman
(Eds), Family Medicine: The maturing of a discipline (pp. 129-157). New York:
Haworth Press.

Culpepper, L., & Franks, P. (1983). Family medicine research. Status at the end of the
first decade. JAMA, 249(1), 63-68.

DeHaven, M. J ., Wilson, G. R., & Murphree, D. D. (1994). Developing a research
program in a community-based department of family medicine: One department's
experience. Family Medicine, 26(5), 303-308.

Diamond, R. M. (1993). Changing priorities and the faculty reward system. In R. M.
Diamond & B. E. Adam (Eds), Recognizing faculty work: reward systems for the
year 2000 (Vol. 81, pp. 5-12). San Francisco: Jossey Bass Publishers.

Dill, D. D. (1986). Research as a scholarly activity: context and culture. In J. W. Creswell
(Ed.), New Directions for Institutional Research (Vol. 13, pp. 7-23). San
Francisco: Jossey-Bass Publishers.

Dillman, D. A. (1978). Mail and telephone surveys: The total design method. New York:
John Wiley & Sons.

Doherty, W. J ., Christianson, C. E., & Sussman, M. B. (1987). Introduction. In W. J.
Doherty & C. E. Christianson & M. B. Sussman (Eds), Family medicine: The
maturing of a discipline (pp. xi-xvi). New York: Haworth Press.

Ellwein, L., Khachab, M., & Waldman, R. (1989). Assessing research productivity:
evaluating journal publication across academic departments. Academic Medicine,
64(Jun), 319-325.

Fairweather, J. S. (1996). Academic values and faculty rewards. In D. E. Finnegan & D.
Webster & Z. F. Gamson (Eds), Faculty and faculty issues in colleges and
universities (2nd ed., pp. 361-378). Needham Heights, MA.: Simon and Schuster
Custom Publishing.

Fairweather, J. S. (1996). Faculty work and public trust: Restoring the value of teaching
and public service in american academic life. Boston: Allyn and Bacon.

179

Fairweather, J. S., & Rhoads, R. A. (1995). Teaching and the faculty role: enhancing the
commitment to instruction in American colleges and Universities. Educational
Evaluation and Policy Analysis, 1 7(2), 179-194.

Fink, A. (1995). The survey kit. Thousand Oaks, CA: Sage Publications.

Finkelstein, M. J. (1982). Faculty colleagueship patterns and research productivity.
Paper presented at the annual meeting of the American Educational Research
Association, New York, NY.

Finkelstein, M. J. (1984). The american academic profession: A synthesis of social
scientific inquiry since world war 11. Columbus, OH.: The Ohio State University
Press.

F inkelstein, M. J ., Seal, R. K., & Schuster, J. H. (1998). The new academic generation: A
profession in transformation. Baltimore: Johns Hopkins University Press.

Flanigan, K., Ballinger, P., Grant, H., Bennett, D., Schiller, M., Waller, K., Jordan, L.,
Gierhart, J ., Brodnik, M., Van, S. L., & et, al. (1988). Research productivity
profile of allied health faculty. Journal of Allied Health, 1 7(May), 87-100.

Fox, M. F. (1983). Publication productivity among scientists. Social Studies of Science,
13(2), 285-305.

Fox, M. F. (1996). Publication, performance, and reward in science and scholarship. In
D. E. Finnegan & D. Webster & Z. F. Gamson (Eds), Faculty and faculty issues

in colleges and universities (2nd ed., pp. 408-428). Needham Heights, MA.:
Simon and Schuster Custom Publishing.

Freeman, J ., Loewe, R., & Benson, J. (1998). Training family medicine faculty to teach
in underserved settings. Family Medicine, 30(3), 168-172.

Friedman, R. H., Alpert, J. J ., & Green, L. A. (1994). Strengthening academic generalist
departments and divisions. Journal of General Internal Medicine, 9(4 Supplement
1), S90-98.

Fry, J ., & Gambrill, E. (1978). Introduction. In J. Fry (Ed.), Scientific foundations of
family medicine. Chicago: William Heinemann Medical Books.

Fulton, O., & Trow, M. (1974). Research activity in American higher education.
Sociology of Education, 4 7(Winter), 29-73.

Gaff, J. G., & Wilson, R. C. (1971). Faculty cultures and interdisciplinary studies.
Journal of Higher Education, 42(3), 186-201.

180

Garr, D. R. (1986). Characteristics and job satisfaction of family physicians in full-time
teaching. Family Medicine, 18(5), 269-273.

Geiger, R. L. (1986). To advance knowledge: the growth of A merican research
universities. New York: Oxford University Press.

Geiger, R. L. (1993). Research and relevant knowledge: American research universities
since World War II. New York: Oxford University Press.

Geyman, J. P. (1971). Family medicine as an academic discipline. Journal of Medical
Education, 46(10), 815-820.

Geyman, J. P. (1977). On entry into phase two in family practice development. Journal of
Family Practice, 4(1), 15.

Geyman, J. P. (1978a). Climate for research in family practice. Journal of Family
Practice, 7(1), 69-74.

Geyman, J. P. (1978b). Family practice in evolution: progress, problems and projections.
New England Journal of Medicine, 298(11), 593-601.

Geyman, J. P. (1978c). On the developing research base in family practice. Journal of
Family Practice, 7(1), 51-52.

Gjerde, C. (1994). Faculty promotion and publication rates in family medicine: 1981
versus 1989. Family Medicine, 26(6), 361-365.

Gjerde, C., Clements, W., & Clements, B. (1982). Publication characteristics of family
practice faculty nominated for academic promotion. Journal of Family Practice,
15(4), 663-666.

Glassick, C. E., Huber, M. T., & Maeroff, G. I. (1997). Scholarship assessed: Evaluation
of the professoriate. San Francisco: Jossey-Bass Publishers.

Glueck, W. F., & Jauch, L. (1975). Sources of research ideas among productive scholars.
Journal of Higher Education, 46(January/February), 103-114.

Gottlieb, B. C. (1968). Processes of socialization in american graduate schools. In R. M.
Pavalko (Ed.), Sociology of Education: A book of readings (pp. 258-270). Itasca,
IL.: F.E. Peacock Publishers, Inc.

Hankey, T. L. (1987). The need for a paradigm to guide family practice research. Family
Practice Research Journal, 6(4), 168-174.

Harrington, M., & Levine, D. (1986). Relationship between faculty characteristics and
research productivity. Journal of Dental Education, 5 0(Sep), 518-525.

181

Hekelman, F. P., Zyzanski, S. J ., & F locke, S. A. (1995). Successfiil and less-successful
research performance of junior faculty. Research in Higher Education, 36(2),
235-255.

Henry, R. (1997). Developing research skills for medical school faculty. Family
Medicine, 29(4), 258-261.

Hillman, B. J., Fajardo, L. L., Witzke, D. B., Irion, M., Cardenas, D., & Fulginiti, J. V.
(1989). Factors influencing the success of academic radiologists in publishing
research. Investigative Radiology, 24(11), 849-854.

Hilton, C., Fisher, W., Lopez, A., & Sanders, C. (1997). A relative-value-based system
for calculating faculty productivity in teaching, research, administration, and
patient care. Academic Medicine, 72(9), 787-793.

Hitchcock, M. A., Anderson, W. A., Stritter, F. T., & Bland, C. J. (1988). Profiles of
family practice faculty development fellowship graduates 1978-1985. Family
Medicine, 20(1), 33-38.

Hitchcock, M. A., Bland, C. J ., Hekelman, F. P., & Blumenthal, M. G. (1995).
Professional networks: The influence of colleagues on the academic success of
faculty. Academic Medicine, 70(12), 1108-1116.

Hitchcock, M. A., & Buck, E. L. (1990). Research roles for family physicians. Family
Medicine, 22(3), 191-195.

Hitchcock, M. A., Lamkin, B. D., Mygdal, W. K., Clarke, C. M., & Clarke, S. O. (1986).
Affective changes in faculty development fellows in family medicine. Journal of
Medical Education, 61(5), 394-403.

Hitchcock, M. A., Stritter, F. T., & Bland, C. J. (1993). Faculty development in the health
professions: conclusions and recommendations. Medical Teacher, 14(4), 295-309.

Holloway, R. L., Marbella, A. M., Townsend, J. M., Tudor, J. M., Tollison, J. W., Saultz,
J. W., & Sherwood, R. A. (1995). Defining the need for faculty in family
medicine: results of a national survey. Family Medicine, 27(2), 98-102.

Holloway, R. L., Wilkerson, L., & Hejduk, G. (1997). Our back pages: faculty

development and the evolution of family medicine. Family Medicine, 29(4), 233-
236.

Hosokawa, M., & Zweig, S. (1990). Future directions in family medicine: results of a
Delphi study. Family Medicine, 22(6), 429-433.

182

Hueston, W. J. (1993a). A comparison of university and community-based family
practice physician educators. Family Medicine, 25(9), 576-579.

Hueston, W. J. (1993b). Factors associated with research efforts of academic family
physicians. Journal of Family Practice, 3 7(1), 44-48.

Inui, T. S., Williams, W. T., Jr., Goode, L., Anderson, R. J ., Bhak, K. N., Forsyth, J. D.,
Hutton, J. J ., Wallace, A. G., & Daugherty, R. M., Jr. (1998). Sustaining the
development of primary care in academic medicine. Working group on sustaining
the development of academic primary care. Academic Medicine, 73(3 ), 245-257.

Jackson, M., & MacInnes, I. (1984). Promotion and tenure in family practice in US
medical schools. Journal of Family Practice, 18(3), 43 5-439.

Jencks, C. (1977). The academic revolution. Chicago: The University of Chicago Press.

Jungnickel, P. W., & Creswell, J. W. (1994). Workplace correlates and scholarly
performance of clinical pharmacy faculty. Research in Higher Education, 35(2),
167-194.

Kassebaum, D. G., & Szenas, P. L. (1993). Rural sources of medical students, and
graduates' choice of rural practice. Academic Medicine, 68(3), 232-236.

Kasten, K. L. (1984). Tenure and merit pay as rewards for research, teaching, and service
at a research university. Journal of Higher Education, 55, 500-514.

Katemdahl, D. A. (1994). Factors chosen by department chairs as important to family
medicine. Family Practice Research Journal, 14(2), 177-181.

Knorr, K., Mittermeir, R., Aicholzer, G., & Waller, G. (1979). Individual publication
productivity as a social position effect in academic and industrial research units.
In F. M. Edwards (Ed.), Scientific Productivity: The Effectiveness of Research
Groups in Six Countries (pp. 55-94). Cambridge: Cambridge University Press.

Kraemer, L., & Lyons, K. (1989). Research productivity of allied health faculty in
academic health centers. Journal of Allied Health, 18(Summer), 349-359.

Krahenbuhl, G. S. (1998). Integrating responsibilities and institutional needs. Change,
18-25.

Kuh, G. G., & Whitt, E. J. (1988). The invisible tapestry: Culture in American colleges

and universities. ASHE-ERIC Higher Education Reports. ( Vol. 1). Washington,
DC: Association for the Study of Higher Education.

183

Ladd, E. C., Jr. (1979). The work experience of American college professors: Some data
and an argument, Faculty Career Development (V 01. 2, pp. 1-4). Washington,
DC: American Association for Higher Education.

Lawrence, J. H., & Blackburn, R. T. (1988). Age as a predictor of faculty productivity:
Three conceptual approaches. Journal of Higher Education, 59(1), 22-38.

Liddle, B., Westergren, A., & Duke, D. (1997). Time allocation and research productivity
among counseling faculty. Psychological Reports, 80(Feb), 339-344.

Long, J. S. (1990). The origins of sex differences in science. Social Forces, 68, 1297-
1315.

Long, J. S., & McGinnis, R. (1981). Organizational context and scientific productivity.
American Sociological Review, 46, 422-442.

Mavis, B. B., & Brocato, J. J. (1998). Virtual discourse: The emergence of DR-ED as a

computer-mediated scholarly communications network for medical educators. The
Journal of Educational Computing Research, 19(1), 53-65.

McGaghie, W. C., Bogdewic, 8., Reid, A., Amdt, J. B., Stritter, F. T., & Frey, J. J.
(1990). Outcomes of a faculty development fellowship in family medicine.
Family Medicine, 22(3), 196-200.

McGee, G. W., & Ford, R. C. (1987). Faculty research productivity and intention to
change positions. Review of Higher Education, 11(1), 1-16.

Megel, M. E. (1985). New faculty in nursing: Socialization and the role of the mentor.
Journal of Nursing Education, 24(7), 303-306.

Merenstein, J. H., & Schulte, J. J. (1990). A residency curriculum for the future. The
STFM Task Force on Residency Curriculum for the Future. Family Medicine,
22(6), 467-473.

Merton, R. (1968). Social theory and social structure ( 2nd ed.). Glencoe, IL.: The Free
Press. ‘

Merton, R. (1973). The sociology of science. Chicago: The University of Chicago Press.
Merton, R. K., Reader, G. G., Kendall, P. L., & (eds). (1957). The student-physician:
Introductory studies in the sociology of medical education. Cambridge, MA.:

Harvard University Press.

Middaugh, M. F. (2001). Understanding faculty productivity: standards and benchmarks
for colleges and universities. San Francisco: Jossey-Bass Publishers.

184

Morzinski, J. A., Diehr, S., Bower, D. J ., & Simpson, D. E. (1996). A descriptive, cross-
sectional study of formal mentoring for faculty. Family Medicine, 28(6), 434-438.

Nettles, M. T., & Pema, L. W. (1995). Sex and race differences in faculty salaries,
tenure, rank and productivity: Why, on average, do women, African Americans,
and Hispanics have lower salaries, tenure, and rank? Paper Presented at the
annual meeting of the Association for the Study of Higher Education.

Neumann, Y. (1979). Research productivity of tenured and non-tenured faculty in US.
Universities: A comparative study of four fields and policy implications. Journal
of Educational Administration.

Nutting, P. A. (1996). Practice-based research networks: building the infrastructure of
primary care research. Journal of Family Practice, 42(2), 199-203.

Oeffrnger, K. C., Roaten, S. P., Jr., Ader, D. N., & Buchanan, R. J. (1997). Support and
rewards for scholarly activity in family medicine: a national survey. Family
Medicine, 29(7), 508-512.

Parkerson, G. R., & et a1. (1982). Meeting the challenge of research in family medicine:
report of the study group on family medicine research. Journal of Family
Practice, 14(1), 105-113.

Pellegrino, E. D. (1987). Family practice facing the twenty-first century: Reflections of
an outsider. In W. J. Doherty & C. E. Christianson & M. B. Sussman (Eds),
Family medicine: The maturing of a discipline (pp. 23-47). New York: Haworth
Press.

Pelz, D. C., & Andrews, F. M. (1966). Scientists in organizations: Productive climates
for research and development. Ann Arbor, MI: Institute for Social Research, The
University of Michigan.

Peterson, M. W., Cameron, K. S., Mets, L. A., Jones, P. a., & Ettington, D. (1986). The
organizational context for teaching and learning: A review of the research
literature. Ann Arbor, MI: National Center for Research to Improve
Postsecondary Teaching and Learning.

Peterson, M. W., & Spencer, M. G. (1990). Understanding academic culture and climate.
In W. G. Tierney (Ed.), Assessingacademic climates and cultures £8 (Winter
1990), (pp. 3-18). San Francisco: Jossey-Bass, Publishers.

Phillips, T. (1978). Research considerations for the family physician. Journal of Family
Practice, 7(Ju1), 121-128.

Print, M., & Hattie, J. (1997). Measuring quality in universities: An approach to
weighting research productivity. Higher Education, 33(4), 453-469.

185

Rebne, D. (1989). The social action perspective: attachments to workload and
productivity in the research function. Educational Administration Quarterly,
25(1), 37-57.

Reskin, B. (1977). Scientific productivity and the reward structure of science. American
Sociological Review, 42, 491-504.

Reynolds, P. P., Giardino, A., Onady, G. M., & Siegler, E. L. (1994). Collaboration in the
preparation of the generalist physician. Journal of General Internal Medicine, 9(4
Supplementl), 855-63.

Rodnick, J. E. (1987). Research in family medicine: counterculture continued. Family
Medicine, 19(4), 251-307.

Rogers, J. (1995). Family medicine research: a matter of values and vision. Family
Medicine, 26(3), 180-181.

Rogers, J. C., Holloway, R. L., & Miller, S. M. (1990). Academic mentoring and family
medicine's research productivity. Family Medicine, 22(3), 186-190.

Schatz, I. J ., Realini, J. P., & Charney, E. (1996). Family practice, internal medicine, and
pediatrics as partners in the education of generalists. Academic Medicine, 71(1),
35-39.

Schein, E. H. (1992). Organizational culture and leadership ( Second ed.). San
Francisco: Jossey-Bass, Publishers.

Scherger, J. E. (1997). Phase three of academic family medicine. Family Medicine, 29(6),
439-440.

Schneid, D. C., Hamm, R. M., & Crawford, S. A. (2000). Measuring academic
production--Caveat inventor. Academic Medicine, 75(10), 993-995.

Sheets, K., & Schwenk, T. (1990). Faculty development for family medicine educators:
An agenda for fiiture activities. Teaching and Learning in Medicine, 2(3), 141-
148.

Snow, C. P. (1959). The two cultures and the scientific revolution. Cambridge:
Cambridge University Press.

Stange, K. C., & Hekelman, F. P. (1990). Mentoring needs and family medicine faculty.
Family Medicine, 22(3), 183-185.

186

Stange, K. C., Miller, W. L., & McWhinney, I. (2000, July 27, 2000). Developing the
knowledge base of family practice. Paper presented at the Keystone 11
Conference, Denver, Colorado.

Stephens, G. G. (1979). Family medicine as counterculture. Family Medicine Teacher
1979, 16(5), 14-18.

Stephens, G. G. (1982). The intellectual basis of family practice. Tucson, AZ.: Winter
Publishing Company.

Stephens, G. G. (1987). Developmental assessment of family practice: An insider's view.
In W. J. Doherty & C. E. Christianson & M. B. Sussman (Eds), Family Medicine:
The maturing of a discipline (pp. 1-15). New York: Haworth Press.

Stephens, G. G. (1990). Family practice in the 1980’s: A second decade of essays. Kansas
City, MO.: Society of Teachers of Family Medicine. ‘

Swee, D. E., Winter, R. O., & Hammond, B. L. (1989). Testing a model of academic
career development. Family Medicine, 21(5), 350-3 54.

Sweeney, R., & Jones, J. G. (1993). The case for family practice and primary care
research at the NIH. American Family Physician, 47(1), 37-39.

Taylor, R. B., Colwill, J. M., Puffer, J. C., Heffron, W. A., Marsland, D. W., Rake], R. E.,
& Geyman, J. P. (1991). Success strategies for departments of family medicine.
Journal of the American Board of Family Practice, 4(6), 427-436.

Thompson, R., Haber, D., Chambers, C., Fanuiel, L., Krohn, K., & Smith, A. J. (1998).
Orientation to community in a family practice residency program. Family
Medicine, 30(1), 24-28.

Tierney, W. G., & Rhoads, R. A. (1993). Enhancing promotion, tenure and beyond:
Faculty socialization as a cultural process ( Vol. 6). Washington, DO: The

George Washington University/School of Education and Human Development.

Van Maanen, J. (1976). Breaking in: Socialization to work. In R. Dubin (Ed.), Handbook
of Work, Organization, and Society. Chicago: Rand McNally.

Vardan, S., Smulyan, H., Mookherjee, S., & Mehrotra, K. G. (1990). Factors encouraging
research productivity in a division of general internal medicine. Academic
Medicine, 65(12), 772-774.

Wagner, R. (1981). The invention of culture. Chicago: The University of Chicago Press.

187

Wanner, R. A., Lewis, L. S., & Gregorio, D. I. (1981). Research productivity in
academia: a comparative study of the sciences, social sciences, and humanities.
Sociology of Education, 54(238-53).

Weidman, J. C., & Stein, E. L. (1990). The professional socialization of ggduate students
in educational administration (ERIC Document Reproduction Service No.ED 326
945).

Wilson, L. (1942). The academic man. New York: Oxford University Press.

Woods, S. E., Reid, A., Amdt, J. E., Curtis, P., & Stritter, F. T. (1997). Collegial
networking and faculty vitality. Family Medicine, 29(1), 45-49.

Xu, G., Veloski, J ., Hojat, M., Rabinowitz, H. K., & Rattner, S. L. (1997). Factors
influencing primary care physicians' choice to practice in medically underserved
areas. Academic Medicine, 72(10 October Supplement 1), S109-111.

Zyzanski, S. J ., Williams, R. L., F locke, S. A., Acheson, L. S., & Kelly, R. B. (1996).

Academic achievement of successful candidates for tenure and promotion to
associate professor. Family Medicine, 28(5), 358-363.

188