This is to certify that the
dissertation entitled

DROPPING OUT OF SCHOOL: ASSESSING THE EFFECT
OF SUPPORTIVE MESSAGES FROM FAMILY ABOUT
STAYING IN SCHOOL

presentedby

Renee Strom

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Communication

 

 

 

I

4! '7 w .
x‘lj. I 4‘
. / fll// [53 46x

u' Major Professor’s Signature

27 (7Lfi€ 7066

Date

MSU is en Aiiirmative Acflonfiquel Oppommlty Institution

 

LIBRARY
Michigan State
University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 pleIRC/DateDue.indd-p.1

 

 

DROPPING OUT OF SCHOOL: ASSESSING THE EFFECT OF SUPPORTIVE
MESSAGES FROM FAMILY ABOUT STAYING IN SCHOOL
By

Renee Strorn

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Communication

2006

ABSTRACT
DROPPING OUT OF SCHOOL: ASSESSING THE EFFECT OF SUPPORTIVE
MESSAGES FROM FAMILY ABOUT STAYING IN SCHOOL
By
Renee Strom
The social support fi'arnework is used to address how the number and valence of

messages in the home about educational achievement may differently affect how students
progress in school. It is hypothesized that students with higher educational attainment
will recall numerous and positive messages emphasizing the importance of education
versus infrequently hearing or advocating the opposite position when compared to those
with lower educational attainment. Additionally, type of support is purported to affect
achievement, so that students who graduated will recall more emotional and instrumental
support messages than will students who did not graduate. Eighty subjects completed an
in-depth interview about their experiences in high school. Perceived helpfulness emerged
as the only statistically significant and substantial predictor of educational attainment. A
post-hoe causal model was created with perceived helpfulness mediating the effect of
instrumental messages and loss-framed messages on educational attainment. Pregnancy,
parental income, gender, sibling dropout, and message repetition appeared to exert direct
effects on educational attainment. Directions for subsequent tests of this model are

discussed.

ACKNOWLEDGMENTS

I would like to thank Dr. Frank Boster for his guidance and patience during the
writing of this dissertation, as well as through my graduate career. I have learned much
from him about the research process and will apply the knowledge I have gained to my
future research endeavors.

I would also like to thank Dr. Sandi Smith for serving on my dissertation
committee. She has been a trustworthy and insightful mentor throughout my graduate
education and I am thankful for all her help.

I would also like to thank Dr. Kelly Monison and Dr. Cass Book for serving on
my dissertation committee. Their willingness to help and flexibility throughout the

process is much appreciated.

TABLE OF CONTENTS

LIST OF TABLES ..................................................................... v
LIST OF FIGURES .................................................................... vi
CHAPTER 1
REVIEW OF THE LITERATURE ................................................... 1
CHAPTER 2
METHOD ................................................................................ 19
Participants ..................................................................... 19
Procedures ..................................................................... 20
Instrumentation ................................................................... 22
CHAPTER 3
RESULTS ............................................................................... 26
CHAPTER 4
DISCUSSION .......................................................................... 38
Appendix A: Informed Consent Form ............................................... 50
Appendix B: Interview Protocol ........................................................ 52
Appendix C: Demographic Questionnaire ......................................... 57
Appendix D: Instructions for Coding and Coding Scheme for
Support Type ................................................................. 59
Appendix E: Instructions for Coding and Coding Scheme for
Kernel State and Framing .............................................. 62
REFERENCES ................................................................................................. 66

iv

LIST OF TABLES

TABLE 1- Descriptive Statistics for Demographic Variables ................... 19
TABLE 2- Correlations for Hypothesized Variables ........................................ 32

TABLE 3- Percentages for gender, parental income, pregnancy,
instrumental messages, loss-framed messages, helpfulness

and message repetition by educational attainment ......................... 34-35
TABLE 4- Means for Sibling Dropout by Educational Attainment ................. 35
TABLE 5- Correlations for Variables in Path Model ....................................... 37

LIST OF FIGURES

FIGURE 1- Post hoc path model depicting the relationship between closeness
and communication patterns ................................................. 36

vi

Chapter 1
Review of the Literature

When adolescents reach the age of 16 they are no longer mandated by the United
States government to stay in school. Research on school dropout has clearly
demonstrated that no single factor, or group of factors, has been identified as the cause of
a decision to leave school before obtaining a high school diploma (Franklin, 1992;
Persuad & Madak, 1992). School dropout is a serious problem because those who fail to
complete school are more likely to be unemployed (Census Bureau, 1999), more likely to
use drugs (Beauvais, Chavez, Oetting, Deffenbacher, & Cornell, 1996), more likely to
need unemployment insurance or welfare (Morris, Pawlovich, & McCall, 1997;
Rumberger, 1987), and more likely to be incarcerated (Bureau of Justice Statistics, 1995).
The causal direction of these relationships remains unclear.

Although a variety of studies have examined the antecedents, correlates, and
consequences of dropping out of school (for reviews see Rumberger 1987, 1995), this
literature is limited as it treats dropouts as a homogenous group. Yet, it is clear that
students leave school for a variety of reasons (e.g., alienation from school, pregnancy,
financial need) and that there is not a single, or simple, profile of a dropout. One
important contribution to school dropout may be differences in supportive messages from
family members about why staying in school is important. Exploration of the differences
among messages about dropping out is needed to help identify effective methods of
dropout prevention and remediation (Lecompte & Goebel, 1987), as current federal
dropout prevention programs have been criticized as ineffective (Dynarski & Gleason,

2002). Despite the United States government’s increased attention concerning students’

K-12th grade education, the high school completion rate for the country has increased
only slightly over the last quarter of a century. Therefore, the purpose of this study is to
assess the potential impact of different types of supportive messages from family on the
decision to stay in, or drop out of, high school.
School Dropout Defined

Research on dropping out must confront a multitude of theoretical and
methodological issues. Central among these problems is disagreement about how to
define and measure dropping out (for a review, see Rumberger, 1987). Many school
systems define dropouts differently, affecting how rates of completion are calculated
(Hammack, 1987). Questions arise when investigators must distinguish between students
who do not complete high school ever versus those who drop out but then return to
school or pursue their Graduate Equivalent Degree (GED). In this study dropout is said to
occur when students leave school without graduating within a four-year period of time,
whether or not they return to school or receive a GED (Fitzpatrick & Yoels, 1992).

In a recent synthesis of research on nonschool correlates of dropout, Rosenthal
(1998) grouped more than 100 variables into 12 constructs (e. g., student, family, and
community contructs that include unalterable demographic characteristics as well as
practices and policies amendable to change). Rosenthal (1998) concluded that the
problem of school dropout cannot be understood in isolation of contextual factors related
to the event. Early school withdrawal is not exclusively an intrapsychic problem for
students; rather, the internal environment of the school as well as external conditions
must be considered (Christenson, Sinclair, Lehr, & Godber, 2001; Jordan, McPartland, &

Lara, 1999). Thus, research on school dropout has clearly demonstrated that no single

factor, or small group of factors, can provide a thorough causal description of the
decision to leave school before obtaining a high school diploma (e.g., Franklin, 1992;
Persuad & Madak, 1992), but instead numerous and varied factors may operate to affect a
student’s decision to stay in, or drop out of, school. A review of the various factors
associated with dropout follows.

Previous Research on Dropout

Traditional research on school dropout has examined demographic predictors of
dropout such as age, socioeconomic status, and ethnicity (e.g., Rumberger, 1983, 1987;
Wilson, 1987). The social factors found to be associated with dropout behavior include
poverty, coming from a single parent family, having parents with a low level of
education, and belonging to a minority language group (Ekstrom, Goertz, Pollack &
Rock, 1986; Fine, 1986; Johnston, Markle & Harshbarger, 1986; McDill, Natriello, &
Pallas, 1985; Rumberger, 1983; Wehlage & Rutter, 1986)} In addition, pregnancy during
high school has been a consistent predictor of dropping out (F orste & Tienda, 1992;
Geronimus & Korenman, 1992; Rumberger, 1983). Further, past research has indicated
that low achievers, males, older students, and working students are more likely than their
peers to drOp out of high school (Battin-Pearson, Newcomb, Abbott, Hill, Catalano, &
Hawkins, 2000; Bickel, 1989; McNeal, 1997; Rumberger, 1987).

Low academic achievement has been the primary aspect of school performance
studied in the last 20 years, and it is typically assessed with standardized achievement
tests or grade point average (GPA). This variable has consistently been one of the
strongest predictors of dropping out of school (Cairns, Cairns, & Neckennan, 1989;

Ekstrom, Goertz, Pollack & Rock, 1986; Fagan & Pabon, 1990; Rumberger, 1983). In

addition, poor early school achievement and grade retention have been found to be
important predictors of high school dropout (Alexander, Entwistle, & Horsey, 1997;
Roderick, 1993). Risk factors include tardiness, chronic absenteeism, truancy,
suspensions, expulsions, retention, academic problems, sense of alienation and
disengagement from school, and poor peer acceptance (Barton, Watkins, & J arjoura,
1997; Cairns, Cairns, & Neckerman, 1989; Ekstrom et al., 1986; Kortering, Hess, &
Braziel, 1997; Sinclair, Christensen, Evelo, & Hurly, 1998). Moreover, dropouts report
being alienated from the social environment in school (Ekstrom et al, 1986), and are
dissatisfied with teacher interest and school discipline (Wehlage & Rutter, 1986).
Students who have dropped out of high school have reported lower occupational goals
and less motivation to learn than those who completed high school (Fine, 1986; Johnston,
Markle, & Harshbarger, 1986). Furthermore, dropouts have reported taking part in fewer
alter-school activities than their graduating counterparts (Ekstrom et a1, 1986; Johnston et
al., 1986; Mahoney & Cairns, 1997; Wehlage & Rutter, 1986). Finally, dropouts are more
likely to have an after school job at which they are employed for more than 15 hours per
week (Fine, 1986; McDill et al., 1985; Rumberger, 1987; Wehlage & Rutter, 1986).
Conversely, school bonding has been linked to academic achievement; students who feel
a sense of attachment to school are more successful academically (Hawkins, Catalano, &
Miller, 1992; Maguin & Loeber, 1996).

Several national studies of educational outcomes have included assessments of the
reasons students leave school prematurely. In a review of the High School and Beyond
dataset, Eckstrom et al. (1986) report school-related factors as the most common reasons

cited for leaving school prematurely (e. g., disliking school, getting poor grades, not

getting along with teachers). Analyses of the High School and Beyond dataset have found
that parental selection factors are related to dropout rates so that dropout rates are higher
in larger schools, in public versus Catholic schools, and in urban schools (Barro &
Kolstad, 1987; Bryk & Thurn, 1989). Additionally, students are more likely to persist to
graduation in schools at which there is an emphasis on academic pursuits and an orderly
environment (Bryk & Thum, 1989). Although research to date has been useful in
identifying the wide range of factors associated with dropping out, many of these factors
are descriptive, and thus reveal little about the underlying processes that actually lead to
dropout behavior. Implicit in much of the research on school dropout is the assumption
that dropping out is a problem of the individual student and that understanding the
characteristics of dropouts will help educators target resources in ways that will reduce
the number of those who fail in the future (Bryk & Thum, 1989). One way to understand
the attitudes and behaviors associated with school dropout is by assessing the types of
interactions students have with their parents about school.
Communication in the Home

Families are often the earliest and most fundamental socialization agents for a
developing child, and socialization outcomes are sometimes dependent on the quality of
parent-child relations. To the extent that children identify with their parents, they are
more likely to internalize their parent’s values (Whitbeck & Gecas, 1988). The way in
which children identify with their parents, in turn, depends on the qualities of the parents
and what type of parental support is exerted. Parental behavior toward the child which is
characterized as supportive and nurturing seems to be most conducive to the child’s

identification with a parent, as well as to a number of other positive outcomes (Maccoby

& Martin, 1983; Rollins & Thomas, 1979). One socialization outcome often addressed in
the research is educational achievement. Luszki and Schmuck (1963) have studied
student perceptions of parental support and interest in their schoolwork; such support
correlates significantly with the student’s own attitudes toward school and performance
in school. Thus, parents who know their child is struggling in school may try to find a
way to communicate with the child about the problem.

Certainly socialization is not the only mechanism that influences a child’s attitude
toward school and subsequently educational achievement. There may be genetic reasons
that explain why a parent and child hold similar views toward school. A study conducted
by Plug (2002) considered the effects of nature (defined as unchangeable genetic
influence) versus nurture (defined as parental educational attainment and aspirations for
schooling) on how parents pass on the importance of education to their children. The
Wisconsin Longitudinal Survey was incorporated which included identification from
adopted and birth children. Plug (2002) found that when heritable ability is controlled,
the nurture component of both mother’s and father’s schooling, and aspirations for
schooling, are the primary factors in explaining schooling differences among both birth
and adopted children. This study is focused on the ‘nurture’ activities of parents by
looking at how variables like parental educational attainment affect school achievement,
while also assessing the effects of specific interactions that occur in the home about
school and school dropout.

Parents influence their children’s academic performance by imparting the values,
aspirations, and motivation needed to persevere and succeed in school (Wright & Wright,

1976). The process of educational attainment involves coordinating decisions across

many dimensions, (e.g., curriculum placement, curriculum choice, participation in
extracurricular activities, and post-secondary school choice). Successfully navigating this
complicated system is partly dependent on parental assistance (Baker & Stevenson,
1986). Rumberger, Ghatak, Poulos, Ritter, and Dombush (1990) found that an important
difference between students who did poorly in school and dropped out and those who did
poorly in school and remained was the extent to which both students and their parents
were involved in the child’s education and educational decisions. Research has also
shown that dropouts tend to come from homes with a weaker educational support system.
Research demonstrates that after controlling for socio-economic status dropouts come
more often from families characterized by a lack of supervision, a permissive parenting
style, low aspirations regarding the schooling of their children, and negative reactions to
school underachievement (e. g., unrealistic expectations that lead to excessively harsh
punishments) (Astone & McLanahan, 1991; Ekstrom et al., 1986; Fagan & Pabon, 1990;
McCombs & Forehand, 1989; Rumberger, 1983, Rumberger et al., 1990). Additionally,
dropouts had fewer study aids present in their homes, had less Opportunity for non-school
related learning, and had mothers with lower levels of formal education (Ekstrom, etal.,
1986). A meta-analysis conducted on communication in the home about dropout (Strom
& Boster, 2005) revealed a direct positive relationship between shared parental
aspirations and school completion (Alexander, Entwistle, & Horsey, 1997; Alexander,
Entwistle, & Kabbani, 2001; Battin-Pearson, Newcomb, et al., 2000; J imerson, Egeland,
Sroufe, & Carlson, 2000; Persuad & Madak, 1992). This effect is consistent with the

proposition that shared parental aspirations may affect high school completion.

Parenting styles that foster increased supportive communication between parents
and their children also appear to influence students’ achievement in school (Dombusch,
Ritter, Leiderman, Roberts, & Fraleigh, 1987). Interaction between parents and children
about school is the most direct way that children will experience their parent’s views on
schooling, because through interaction they must confront the others beliefs and reconcile
them with their own, if necessary. Previous research has found that as parent-child
interactions about school increase, school completion increases (Hess & Copeland, 2001;
J anosz, LeBlarIc, Boulerice, & Tremblay, 1997; Vallerand, Fortier, & Guay, 1997). A
longitudinal study by Ensinger and Slusarcick (1992) found that by age 16 if the mother
reported that the adolescent confided about school, worked on homework with adult
family members, and sought advice from the parent(s) about school, the adolescent was
more likely to graduate from high school. In addition, negative family interaction has
been a risk factor for drop out that has received more attention in recent research (Barton
etal., 1997; Cairns et al., 1989; Kortering, Hess, & Braziel, 1997; Sinclair, Christenson,
Evelo, & Hurly, 1998). Thus, it is possible that the types of messages a parent gives to a
child about school (e.g., “You must do well in school to be successful in your life”) may
affect how a child evaluates whether to stay in, or drop out of, school.

The decision to drop out usually develops over a period of time (e. g.,
disengagement from school over the course of several years due to poor grades, repeating
a grade, lack of participation in extracurricular activities) and is infrequently an impulsive
decision (Roderick, 1993). The type of communication that is occurring in the home
about educational achievement may serve as an important influence on a student who

may be thinking about dropping out (e. g., “If my own parent doesn’t care if I stay in

school, why should 1?”). The number of messages sent and the valence of the messages
(e.g., positive or negative) may differentially affect how a student assesses the
importance of graduating from high school. Social support is one area of study that has
addressed positive forms of communication in personal relationships like the family.
Social Support

Social support provided in interpersonal relationships has been found to be
emotionally beneficial (see Cohen & Wills, 1985 for a review) and has been found to
influence psychological well-being (e.g., Billings & Moos, 1982; Cobb, 1976) as well as
physical health (e.g., Antonucci & Jackson, 1987; Shwarzer & Leppin, 1991). Some
socially supportive interactions are related to a variety of positive outcomes including
reducing stress, enhancing self-esteem, and providing tangible assistance with stressful
experiences (for a review see Krause, 1990). Receiving support from others has been
shown to moderate the effects of daily challenges and disappointments that, if
unattended, can grow into major stressors (e.g., Eckenrode, 1984; Stone & Neale, 1984).
Social support is defined as “information leading the subject to believe he/she is cared for
and loved, esteemed, and a member of a network of mutual obligations” (Cobb, 1976, p.
300). Both short-term and long-term supports are likely to consist of help mobilizing
psychological resources and mastering emotional burdens, sharing tasks, and providing
extra supplies (e.g., tools, skills) to improve the handling of a situation (Caplan, 1976).

An important distinction to note concerning how social support is conceptualized
is that of received support versus perceived support. Received support is support
participants report having had and is assessed by self-report measures that were designed

to obtain participants’ perceptions of the amount of support they have received in the

recent past. Perceived support is the perception that support is available if needed in the
future (Cohen & Willis, 1985). Instruments designed to measure perceived support have
respondents indicate the likely availability to them of various types of support, should the
need arise. This study addresses received support, as subjects will be asked to recall
support messages from primary caregivers.

It is logical for social support to be studied as communication because it is
ultimately conveyed through messages directed by one individual to another in the
context of a relationship that is created and sustained through interaction (Burleson &
MacGeorge, 2002). The study of social support is the study of supportive communication
through verbal and nonverbal behaviors intended to provide help. Research incorporating
social support as communication involves studying the messages through which people
both seek and express support, studying the interactions in which supportive messages
are produced and interpreted, and studying the relationships that are created by the
supportive interactions in which people engage (for a review see Burleson & MacGeorge,
2002)

It is important to consider that even the most well-intentioned support attempts
may fail because they are excessive or untimely (Wortman & Lehman, 1985). Social
relations that we rely on for support may have costs as well as rewards even when
support attempts are well intentioned. Social support only takes into account messages
that make people feel esteemed and loved, not those that make people feel unworthy and
unloved (Dunkel-Schetter, Blasbland, F einstein, & Herbert, 1992). In contrast, people
involved with a person in distress, particularly those closest to that person, may become

critical and hostile to the stressed person and become psychologically distressed

10

themselves (Coyne, Wortman, & Lehman, 1988) which may lead to a lack of positive
supportive attempts. Andrews, Tennant, Hewson, and Schonell (1978) found that the
combination of stressful life events, low level of social support, and adverse childhood
experiences predicted the occurrence of maladjustment in adults. In addition, negativity
in social interactions has been shown to be related to poorer mental health among stressed
as well as nonstressed populations (e.g., Abbey, Abramis, & Caplan, 1985; Barrera,

1988; Vinokur & van Ryn, 1993). Support providers may become too focused on the
outcomes of their helping efforts to be aware of what they are communicating to the
recipient (e.g., repeatedly telling a person what they should do to fix a problem instead of
just listening may be viewed as negative) or may not be focused enough which can be
viewed as a lack of support.

Thus, it seems students who receive information that make them feel cared for
and loved at home have an advantage over those who do not receive supportive messages
in the home. This advantage may occur because supportive information has a booster
effect in which the effect of social support enhances the beneficial impact of a positive
life event like doing well in school. Conversely, supportive information received in the
home may serve as a buffering effect, namely reducing the adverse impact of negative
school experiences. Previous research has documented how social support served as a
booster for positive life events (Murrell & Norris, 1983), as well as the stress-buffering
effect of social support (Cohen & Wills, 1985; Kessler & McLeod, 1985). Research
specific to educational attainment has provided evidence for family support messages
serving as a buffer against negative school events (Okun, Sandler, and Baumann, 1988).

Sandler (1981) investigated how social support affected students' satisfaction with their

11

college experience and found evidence for boosting (i.e., enhancing the positive) and
buffering (i.e., reducing the negative) effects of social support on satisfaction with
college. Such effects may be easily applicable to high school students. Take, for instance,
the case of the student who is doing very well in school. Such a student may have a high
GPA, may be involved in school, and may never have struggled with grades. If this
student receives positive messages in the home about school achievement and is doing
well in school a booster effect may occur (e.g., positive social support messages are
reinforcing positive school achievement) thus promoting continued achievement in
school. Conversely, positive messages received by a student who has been struggling in
school (e.g., being bullied, does not like a teacher, is failing a class) may serve as a
buffering effect against such negative school experiences (e.g., alleviate feelings of
inadequacy or embarrassment), thus allowing the student to move past these problems
and continue in school. Sometimes negative school experiences may serve to help
students realize their limitations (e.g., continually getting reprimanded by a teacher for
bad behavior in class). In this instance a positive message from a parent may take the
form of the parent explaining to the child why the teacher is reprimanding the student and
explain why it is important to pay attention. Therefore, a message may be positive or
negative depending on how it is communicated. In this example the message is the same,
but is delivered differently and thus may be received differently by the student. Thus, it is
plausible that students who perceive hearing more support messages will be more likely
to have higher educational attainment due to either a booster and/or buffer effect of
positive social support. An increase in positive messages may produce an enhanced

booster effect (if the student is already experiencing positive events at school) or may

12

produce a strong buffer effect against the perception of negative events that are occurring
at school, thus reducing the impact of such events. Therefore the first hypothesis in this
study is that students with higher educational attainment will recall numerous and
positive messages emphasizing the importance of education versus infrequently hearing
or advocating the opposite position when compared to those with lower educational
attainment.
Different Types of Social Support

Social support has been categorized in various ways (Beehr, 1985; House, 1981;
LaRocco, House, & French, 1980). Two types of social support, emotional and
instrumental support, have received the most attention in the support literature:
Emotional support consists of expressions of care, concern, sympathy, understanding,
encouragement, and indicates acceptance of a person (Dunkel-Schetter, Blasbland,
Feinstein,& Herbert, 1992). In contrast, instrumental support is viewed as rendering
tangible assistance, such as physical assistance or aid in the form of advice or guidance
(Kaufinann & Beehr, 1986). An important criterion for assessing the probable efficacy of
supportive communication is if support recipients find supportive attempts helpful. For
example, although a message conveying advice may present a perfect solution to
another’s problem, such advice may not be effective if recipients fail to heed it because
they feel threatened or offended. Thus, it is important to understand the perceived
helpfulness of different types of support messages.

Messages are experienced as providing firnctional emotional support when
helpers clearly express the desire to be helpful or provide support; express acceptance,

love, positive regard, and affection for the target; express concern, care, and interest

13

about the target’s current situation; express their availability to the target; and express
alliance with the target (Burleson, 2003). Messages coded as conveying emotional
support have regularly been identified by participants as either the most helpful or one of
the most helpful forms of assistance they recall receiving from others (Barbee, Derlega,
et al., 1998; Caplan & Samter, 1999; Cramer, 1990; Dakof & Taylor, 1990; Dunkel-
Schetter, 1984; Goldsmith, 1994; Sullivan, 1996). In addition, messages that legitimize
the target’s feelings and actions, and that express acknowledgement, comprehension, and
understanding of the target’s feelings and situation are also experienced as helpful
(Burleson, 1994b; Burleson & Samter, 1985a; Samter, Burleson, & Murphy, 1987).
Additional benefits of emotional support include alteration of threat appraisal of life
events, enhanced self-esteem, reduced anxiety/depression, and motivated coping (Willis
& Shriner, 2000).

Instrumental support attempts may include offers of information about resources,
suggest alternative courses of action, and provide advice about effectiveness. Benefits
may include an increase in the amount of useful information available to the support
recipient, help in obtaining needed services, and may lead to more effective coping
(Willis & Shriner, 2000). Perceived helpfulness of informational support attempts is
contingent on how the question is asked. Sharing information about the problem is
experienced as helpful, provided the target sees the information as pertinent to the
problem (Cutrona, 1990; Cutrona, Suhr & McFarlane, 1990). The perceived helpfulness
of one type of instrumental support message, providing advice, has been called into
question. Research has found that providing advice about how to manage a problem has

been viewed as unhelpful (Chesler & Barbarin, 1984; Dakof and Taylor, 1990; Dunkel-

l4

Schetter, 1984), helpful (Cutrona, Suhr & McFarland, 1990; Sullivan, 1996), or has failed
to have an effect (Cutrona & Suhr, 1994). The relationship between advice and perceived
helpfulness may be unclear because there may be important moderators influencing the
relationship. According to Cutrona and Russell (1990) different types of stressful events
call for different forms of coping and, consequently, different forms of social support will
serve as an optimal match for the needs of the distressed person. One dimension that may
affect the perceived helpfulness of advice messages is the controllability of the situation,
where advice (and other forms of information and guidance) will be most helpful in
challenging or threatening situations in which negative consequences can be prevented or
altered (Cutrona & Suhr, 1994). Despite the mixed results on advice giving, the role of
instrumental support is to provide guidance and alternatives for problematic life events,
which clearly provides a different type of support than making a person feel loved and
esteemed.

It is important to note that many of the studies that have assessed the effectiveness
of emotional support versus instrumental support have examined how supportive
messages have affected people with health problems like cancer (e. g., Chesler &
Barborin, 1984; Dakof & Taylor, 1990; Dunkel-Schetter, 1984), HIV (e.g., Barbee,
Derlega, Sherbume, & Grimshaw, 1998), or multiple sclerosis (Lehman & Hemphill,
1990). Other studies have looked at how support messages have influenced stress in
personal relationships like marriage (e.g., Cutrona & Suhr, 1994; Sprecher, Metts,
Burleson, Hatfield, & Thompson, 1995) or close friendships (Wilson & Kunkel, 2000).
The type of stressful situation involved influences the type of support that is viewed as

helpful or unhelpful by support recipients (for a review see Cohen & Wills, 1985). Only a

15

handful of studies have assessed the link between perceived social support and
performance on academic or academic-like tasks. Subjects high in perceived social
support have been shown to perform better on a difficult anagram test and report less
cognitive interference, concentrate more on performing the task, and report fewer
interfering thoughts and worries during the task than subjects low in social support
(Sarason, Sarason, Keefe, Hayes & Shearin, 1986). Rosenfeld, Richman, and Bowen
(2000) found that students who reported receiving low support from their parents, fiiends,
and teachers had the poorest school outcomes (poor attendance, problem behavior, low
school satisfaction, engagement, self-efficacy, and obtained poor grades). In addition,
Cutrona, Cole, Colangelo, Assouline, and Russell (1994) found parental social support to
be a significant predictor of college grade point average in two independent samples. It is
important to note that much of the literature on the effects of social support on school
achievement have used global measures of social support and have not distinguished
between how different types of support may differentially impact school achievement. In
addition, there is a dearth of studies that have assessed how different types of support
may differentially irnpact school achievement at the high school level, thus claims of
perceived helpfirlness of emotional versus informational support types may be different
than in the aforementioned contexts. Yet understanding different support attempts may
provide some answers concerning why some students reach higher levels of educational
attainment than others. One area of study that closely approximates stress in school is
research conducted in organizations in which social support functions to alleviate job ,
stress (e.g., Kaufinann & Beehr, 1986), as attending school can be much like attending a

job for many students. Much of the research on job stress has concentrated on the effects

16

of emotional support, apparently assuming that it is the most important type (Chisholm,
Kasl & Mueller, 1986; LaRocco et al., 1980). Yet recent research on support in
organizations has stressed the importance of assessing the effects of both emotional and
instrumental support on important outcomes as both types may play important, albeit
different, roles in affecting job outcomes (F enlason & Beehr, 1994).

A student who is struggling in school may benefit from emotional support
messages (e.g., parents tell struggling students they are accepted and loved no matter
what is going on at school) as well as instrumental support messages (e. g. parent offers to
help the student with homework, talk to school officials about what the student can do to
improve, give advice about the best way to study or handle a bully at school). Both types
of support messages may provide different, but equally important, forms of knowledge
that students can use to alleviate anxiety as well as solve an immediate problem (e. g., I
know am loved no matter what happens at school, and I know I can work to address this
problem by studying harder for tests). But receiving emotional support only may not be a
sufficient form of support in such a decision-making process. Students may also need
supportive information about actions they can take to solve or ease the problem in order
to make a more informed decision concerning school. Thus, greater emotional support
(i.e., acceptance, encouragement, and praise) and instrumental support (i.e., advice,
guidance) may buffer or reduce the negative psychological consequences that increased
school roles, demands, and conflicts can have on school achievement. Thus, both
emotional and instrumental support types may be crucial in a student’s decision-making
process concerning educational achievement, where students who receive more emotional

and instrumental support messages will have higher educational achievement. Thus, the

17

second hypothesis posited is that students with higher educational attainment will recall
more emotional support messages about educational achievement than will students with
lower educational attainment. The third hypothesis posited is that students with higher
educational attainment will recall more instrumental support messages about educational
achievement than will students with lower educational attaimnent.

This study, therefore, will explore if the number and valence of support messages,
as well as the type of support messages, a parent gives about school will affect a person’s

decision about staying in, or dropping out of, high school.

18

Chapter 2
Method

Participants

Eighty subjects (48 females and 32 males) participated in the present study. Their
ages ranged from 18 to 38, M = 25.7 (SD = 5.56) (see table 1 for descriptive statistics).
Their self-reported ethnicity was Asian (3.7%), Black (12.3%), Hispanic (9.9%),
Caucasian/White (66.7%), and Bi/Multi-racial (6.3%). Participants were recruited
through a snowball sampling technique. This technique uses a process of chain referral
where members of the target population are located and are asked to provide names and
addresses of other members of the target population, who are then contacted and asked to
name others (Singleton & Straits, 1999). This sampling technique was designed to obtain
equal numbers of the following four educational attainment categories: high school
dropouts with no further education, high school dropouts who have obtained (or are
working toward) their GED, high school graduates with no college, and high school

graduates with a college degree (or who are working towards one).

Tablel

Descriptive Statistics for Demographic Variables

 

 

 

Variables M SD Rangea
Age 25.7 5.56 18-38
Years in High School 3.15 1.17 0-5
GPA 2.62 .94 0-4
Note. N = 80

19

Subjects were also asked to report on the educational achievement for their father,
mother and for themselves. Sixty-nine subjects were able to report on their father’s
educational achievement (3% had no high school, 23% had some high school, 29%
received their high school degree, 17% had some college, 4% received a 2 year college
degree, 15% received their bachelor’s degree, and 9% attained a Masters degree).
Seventy-six subjects were able to report on their mother’s educational achievement (4%
had no high school, 21% had some high school, 28% received their high school degree,
17% had some college, 9% received a 2 year college degree, 15% received their
bachelor’s degree, and 6% attained a Masters degree). All subjects were able to report on
their own educational achievement (1.3% had no high school, 43.8% had some high
school, 3.8% reported receiving their GED, 25% received their high school degree,
16.3% had some college, 3.8% received a 2 year college degree, and 6.3% received their
bachelor’s degree). On average subjects spent a little over 3 years in high school (M =
3.15, SD =1.17).

Socioeconomic status of participants was assessed through participant reports of
their income today as well as the income of their primary giver(s) while they lived with
them in high school. Sixty-five percent of participants reported an income of $15,000 or
below, 18% between $16,000-25,000, 11% between $26,000-35,000, 4% between
$36,000-50,000, and 3% made $50,000 or above. Sixteen percent of participants reported
that their primary caregiver(s) made between $15,000 or below while living with them,
20% between $16,000-25,000, 16% between $26,000-35,000, 15% between $36,000-
50,000, and 33% reported $50,000 or above.

Procedure

20

The investigator conducted audio taped, face-to-face interviews. Subjects were
asked a series of close-ended and open-ended questions about their high school
experiences as well as about messages they recalled receiving high school about school
achievement. Prior to each interview, participants completed an informed consent form
(see Appendix A), and their confidentiality was guaranteed as their name was not used in
conjunction with their responses. All interviews were tape-recorded but no identifying
information was associated with any of the tapes. Prior to conducting the interviews,
several mock interviews were conducted to refine the interview protocol. Interviews
generally lasted 25-30 minutes, and all subjects were able to successfully complete their
interview. At the conclusion of each interview subjects were thanked, debriefed, and
were given $15 for their participation.

A funneling technique was implemented in the interview to help respondents
remember high school experiences. This technique was incorporated by beginning the
interview with broad questions about the subject’s high school experiences. As the
interviews continued the questions became more specific to certain aspects of high
school. This strategy allowed the subject to remember back to high school without asking
questions that may have been difficult to answer initially. As the interview continued the
questions became more specific and were easier for subjects to answer because subjects
were already in a general frame of mind about high school.

The tape-recorded interview was divided into five sections. The first section of
the interview assessed overall context while in high school and was used to help
respondents remember what happened in high school (e.g., I would like you to think

about when you were in high school and try to remember as many aspects of high school

21

that you can. Think specifically about a period of time when you may have been
struggling in school and/or a time when you were doing exceptionally well in high
school. Can you think of a time like this?) The second section of the interview assessed
specific aspects of the school context (e. g., Where did you go to high school? Did you
attend high school anywhere else? Did you like school, dislike school, both like or
dislike, or neither like nor dislike school while attending it?). The third section of the tape
recorded interview assessed farme context (e. g., While in high school did you live with
both parents? Did you live with one parent? Do you have any brothers and sisters?) The
fourth section of the interview asked specific questions about memorable messages
respondents recalled receiving while in high school. Once respondents recalled the
messages a series of questions were asked about the messages (e.g., Who said this to
you? Did you ask for this information? Did your mom/dad/primary caregiver tell you this
a lot?). Subjects were then asked about peers they spent time with in high school, the fifth
and final section of the recorded interview (e.g., Did you have fiiends in high school who
dropped out of high school? How many dropped out? How did that affect you?) (see
Appendix B for entire interview protocol).

At the conclusion of the recorded interview, respondents completed a number of
demographic items (e.g., age, income) (See Appendix C for items).
Instrumentation

Number and Valence of Support Messages. In order to determine number of
support messages, respondents were asked to recall all messages about educational
achievement they received from a primary caregiver while in high school. This procedure

is similar to one used successfully by Smith and Ellis (2001) about the recall of

22

memorable messages. The number of messages recalled was used to investigate the first
and second hypotheses. In order to assess the valence of each message respondents were
asked to answer questions concerning how positive or negative they viewed the support
message (e. g., The first time someone said this to you did you take it as a positive
message about education, a negative message about education, or as a neither positive nor
negative message about education? Somewhat positive, very positive? Somewhat
negative, very negative?) Along with positive versus negative nature of the message,
subjects were also asked to report on the helpfulness of the message (yes or no) as well as
if the message encouraged or discouraged school pursuits. The three measures were
combined to determine if they created a one-dimensional measure of message valence.
The items were highly correlated but did not pass the test of parallelism. Thus the items
were included in the analysis as separate predictors of educational attainment.

Type of Support. In order to investigate the third and fourth hypotheses all
messages were coded as emotional and/or instrumental support type. According to
Dunkel-Schetter et al. (1992) three different perspectives are possible for coding support
messages: the message source or helper, the message recipient or target, and third-party
observers (trained experts). The third option was incorporated in this study. Two coders
were trained to code each message according to the naturalistic paradigm (Dunkel-
Schetter et al., 1992). Participants in such research provide retrospective self-reports of
messages they have received fiom others. The messages are then coded within several
broad support categories (see Clark & Stephens, 1996 for an example). In the present
study all messages were coded as one of four possible support codes: 1) instrumental, 2)

emotional, 3) both instrumental and emotional, or 4) as neither instrumental nor

23

emotional support. Coders were given a list of category codes with examples and were
asked to categorize several messages supplied by the researcher (see Appendix D for
coding scheme). When coders felt comfortable with the system they were asked to code
independently the messages.

In addition to support type, coders also coded each message for kernel state and
framing of each message. Although there were no formal hypotheses posited in this study
concerning the kernel state and framing of the messages, such coding allows for the
investigation of how the wording of the messages may affect educational attainment. A
kernel state is the basic root state mentioned in a message’s description of the
consequence under discussion. Coders were instructed to code the kernel state of each
message as l) desirable (e.g., graduating from high school, going to college, finishing
school); 2) undesirable (not graduating from HS, dropping out, not finishing school); 3)
both desirable and undesirable kernel states (mentions both graduating fi'om high school
AND dropping out of high school) or 4) indeterminate due to phrasing (does not mention
any of the above) (see O’Keefe & Jensen, 2005 for more information on the coding
scheme). Coders were then instructed to code each message for how it was fiamed. Gain-
fiamed messages incorporate an appeal that emphasizes the desirable consequences
associated with compliance with the advocated viewpoint or kernel state. Loss-flamed
messages incorporate undesirable consequences associated with noncompliance of the
advocated viewpoint or kernel state (O’Keefe & Jensen, 2005). Codes for fi'aming
included 1) positive flaming; 2) negative framing; 3) both positive and negative framing;

and 4) indeterminate due to phrasing (see Appendix E for all coding instructions).

24

Educational Attainment. Subjects were asked to report on their level of
educational attainment thus far in their life. Subjects were put into one of four
educational attainment categories: high school dropouts with no further education, high
school dropouts who have obtained (or are working toward) their GED, high school
graduates with no college, and high school graduates with a college degree (or are
working towards one). In the subsequent analysis educational attainment is treated as a
continuous variable in which subjects who graduated from high school and went to
college are treated as having attained the highest level of educational attainment and
dropouts without a GED are treated as having attained the lowest level of educational

attainment

25

Chapter 3
Results
Descriptive Statistics

Along with the demographic questions subjects reported on a series of items about
their experiences with their family, school, and fiiends. The sample is described on the
following dimensions: school variables, family variables, peer variables, and message
variables.

School Context Variables

High School Makeup. Subjects reported attending a variety of high schools.
Thirty-six percent of subjects attended one of the major high schools in the Lansing
School District (i.e., Sexton, Everett, Eastern) and 64% attended other high schools. The
average class size of the schools attended was 345 students. The majority of subjects
attended only one high school (57.5%) dming their high school years. In addition, more
than one-third of subjects (36.3%) reported that they had been retained, or held back a
grade, at some point in their schooling (kindergarten through high school).

Feelings about School. Subjects were asked to report on their feelings about
school while attending high school. Most reported liking school or both liking and
disliking school while attending it (81%). Subjects were also asked if their feelings
toward high school were different now than when they attended. A majority of subjects
reported feeling more positive about school now (73%), and few reported feeling more
negative (7%). The average GPA for subjects was 2.62 (SD = .94). Subjects’ ratings of
their academic performance in school were varied with 43% reporting they did well and

30% reporting they did poorly in school.

26

Time Spent Outside of School. Subjects were asked if they held a job during the
school year, and slightly more than one-half reported working during the school week
(60%). Those who worked during high school reported working less than 15 hours per
week on average (M = 13.15, SD = 12.03), which is considered part-time work. Subjects
also reported on their involvement in extracurricular activities during their high school,
and a majority of subjects participated in one or more activities that took place outside of
school hours but during the school year (60%). On average those involved spent an
average of 6.2 participation hours per week (SD = 6.07).

Pregnancy. Subjects were asked if there had been a pregnancy that had
contributed to how they performed in school. Few participants reported being pregnant,
or getting someone pregnant, in high school (12.5%). All reports of pregnancy were
linked to reports of a time subjects struggled in school.

Feelings toward Teachers. Subjects were asked if there were specific teachers in
high school they liked. Most participants were able to recall at least one teacher they
liked (84%). A substantial percentage reported having at least one teacher they disliked in
school (69%).

Family Context

Family Life. A number of questions were asked in order to ascertain the living
situation for subjects. Almost one-half of all subjects reported living with both parents in
high school (45%). For those subjects who lived with one parent most reported living
with their mother (41%); only 9% reported living with their father.

Siblings. Almost all subjects reported they had a sibling (95%). On average

subjects reported having three siblings (M = 3.18, SD = 2.41), and a majority of subjects

27

reported that their siblings were biological (61%). Subjects were also asked how many of
their siblings dropped out of high school. On average, subjects reported that they had at
least one sibling who dropped out of high school (M = 1.05, SD = 1.56). One-half of all
subjects’ siblings did not go to college and almost one-third of subjects reported that all
of their siblings went on to college (29%). A little over one-half of subjects reported that
they were close with at least one of their siblings (54%) (e.g., shared important
information) while in high school.

Parents Work Habits. Nearly all subjects reported that one or both of their parents
worked while they were in high school (99%). Similarly, of the working parents almost
all worked fulltime (considered to be 40 hours or more a week) (96%). Subjects were also
asked if their parent(s) liked work. Most subjects felt their parents did like their job or
both liked and disliked their job (88%).

Parental Participation. Subjects were asked if their performance in school was
important to their parents. Most subjects reported that their performance in school was
important to their parents (86%). Almost all subjects reported receiving help with
homework at least once in a while (82%) and only a few reported receiving help daily
(3%). Subjects were also asked to report if their primary caregiver attended parent-
teacher conferences, and most parents did attend (73%). Most subjects reported that their
parent had perfect attendance at conferences (76%).

Peer Context

Friend Dropout. Subjects were asked a series of questions about their high school

fiiends. Subjects were asked if they had fiiends who dropped of out high school, and a

majority of subjects reported having at least one fiiend who dropped out (66%). The

28

average number of fiiends reported to have dropped out was 4.12, SD = 5.81. Of the
subjects with fiiends who dropped out a little more than one-half reported being affected
by their friend(s) dropping out (57%). Nearly all of subjects reported having fiiends who
graduated fi'om high school (96%).

Friend Involvement. Subjects were asked to report if their fiiends were involved
in extracurricular activities and academics in high school. More than one-half of subjects
reported that their fiiends were either somewhat or very involved in activities (68%), and
most subjects reported that some or most of their friends did well academically (86%).

Skipping Habits of Friends. Subjects were asked if their fiiends skipped school
while in high school. More than one-half of subjects reported that their friends skipped
school very often (55%) and only a few reported that their fiiends never skipped school
(11%).

Educational Attainment of Friends. Finally, subjects were asked if they had
fiiends who went on to college. A majority of subjects had friends who went on to
complete some college (63%).

Message Variables

Memorable Messages. Subjects were asked to recall messages they remembered
receiving about educational achievement while in high school. On average subjects
recalled 1.3 messages, SD = .50. A majority of subjects recalled one message (69%).
Only one subject recalled three messages. There were 106 total messages recalled by
subjects.

Coding Messages. Two independent raters coded participants’ messages. Each

complete message recalled by participants served as the unit of analysis. Coders assigned

29

each message a single number to represent one of the four possible support codes:
instrumental, emotional, both instrumental and emotional, or as neither instrumental or
emotional support. The four support codes were then converted into two dichotomous
variables: Emotional support (present or absent) and instrumental support (present or
absent). Cohen’s kappa was used to assess the proportion of agreement between raters
after accounting for chance (Cohen, 1960), Cohen’s kappa = 0.57. Landis and Koch
(1977) have suggested that a kappa coefficient that falls between 0.41 and 0.60 is
considered moderate overall agreement. Discrepant codes were handled by having both
coders discuss for a mutually agreed upon code and this consensus decision constituted
the final code. Very few messages were coded as emotional support (8% of the 106
messages), while 79% of the messages were coded as instrumental support messages.

The same two coders were used to code the kernel state (Cohen’s kappa = 0.76)
and flaming (Cohen’s kappa = 0.77) of each message. According to Landis and Koch
(1977), a kappa coefficient that falls between 0.61 and 0.80 is considered substantial
overall agreement. Discrepant codes were handled by having both coders discuss for a
mutually agreed upon code, and this consensus decision constituted the final code. A
majority of the messages were viewed as having a desirable kernel state (70%), and 28%
of the messages were coded as having an undesirable kernel state. Thirty-five percent of
the messages were coded has having a positive flame, and 31% percent of messages were
coded as having a negative flame.

Message Interaction. Subjects were asked a series of questions that assessed the

interaction surrounding their recalled messages. Nearly one-half of subjects named their

30

mother as the source of their recalled message (49%). One-fourth of the subjects named
their father as the source (24%) and only 15% named both parents as the source.

Message Valence. Subjects were asked three questions concerning the valence
of the message (i.e., Did you see this message as positive or negative? Did this message
discourage you fi'om pursuing school, encourage you to try harder, or neither? Was the
message helpful?) A majority of the messages were described as either somewhat or very
positive (52%) whereas approximately one-fourth of the messages were viewed as neither
positive nor negative (27%). Very few subjects saw their messages as both positive and
negative (3%). In addition, more than one-half of all subjects reported that the message
encouraged them to try harder in school (53%) whereas only a few subjects reported
feeling discouraged by the message (12%). Finally, a majority of subjects also thought
the message was helpful (61%).

Message Repetition. Subjects were asked to report how often they remembered
hearing the message. A little less than one-half of the subjects reported hearing the
message once in awhile (42%) and only a few subjects recalled hearing it on a daily basis
(18%). When asked if the message changed over time, a majority of subjects reported the
message did not change over time (59%), whereas one-fourth of the sample reported the
message became more positive over time (25%).

Message Response. Most subjects reported that they did not ask for the
information included in the message (75%) and less than one-half stated that the message
came in response to something the subject said (43%). A majority of subjects had no
reaction to hearing the message (59%), and almost one-third of the sample reported

reacting in a positive manner to the message (30%).

31

ijothesis Testing

The first hypothesis predicted that students with higher educational attainment
would recall numerous and positive messages emphasizing the importance of education
versus infl'equently hearing or advocating the opposite position when compared to those
with lower educational attainment. To test this hypothesis educational attainment was
first regressed on the number of messages people recalled. This effect was modest as
there was little variance in the number of messages people recalled. To examine the
possibility that the extent to which messages were positive had an impact, educational
attainment was regressed on the valence of the message, whether the message encouraged
or discouraged pursuits in school, and the helpfulness of the message. Perceived
helpfulness emerged as the only statistically significant and substantial predictor of
educational attainment. (r = .45), t (78) = 4.45, p < .01 . Thus, the data are consistent with
the proposition that messages perceived as more helpful have a salutary impact on

educational attainment (see Table 2 for correlations among this set of variables).

Table 2

Correlations for Hypothesized Variables

 

 

Variable l 2 3 4 5
# of Messages

+ or - Valence .06

Discourage/encourage -.05 .60"

Perceived Helpfulness .02 .55" .5 5'”
Educational Attainment .15 .09 .29" .45”

 

Note. N = 80. " p < .05

32

Hypothesis 2 posited that students with higher educational attainment would
recall more emotional support messages about educational achievement than would
students with lower educational attainment. No relationship emerged between emotional
support messages and educational attainment as emotional support type was not
correlated with educational attainment, r = .04, n.s. Thus, hypothesis 2 was inconsistent
with the data

Hypothesis 3 posited that students with higher educational attainment would
recall more instrumental support messages about educational achievement than would
students with lower educational attainment. There does not appear to be a systematic
relationship between instrumental support type and educational attainment as they are not
highly correlated, r = .14, n.s. Thus, hypothesis 3 was inconsistent with the data.
Additional Analyses

There are a number of variables that have conceptual properties that may be
important predictors of educational attainment. Of the variables hypothesized to affect
educational attainment in this study perceived helpfulness is the only one that emerges as
an important predictor, yet it is important to control for other causes of educational
attainment as the effect of perceived helpfulness may be spurious with respect to these
other known causes. Thus additional analyses were performed to tease out the
relationship between perceived helpfulness and educational attainment while controlling
for variables that have previously been found to be important predictors. Alter analyzing
a variety of variables (e.g., parental income, pregnancy, retention), six predictors
emerged as important to educational attainment. When these variables were added (with

perceived helpfulness) to the regression equation, each was found to be a statistically

33

significant and important predictor: parental income, B = .37, t (69) = 4.90, p < .05;
pregnancy, B = -.22, t (69) = -3.1 1, p < .05; gender, B = .25, t (69) = 3.48, p < .05; sibling
dropout, B = -.19, t (69) = -2.55, p < .05; message repetition, B = -.23, t (69) = -3.24, p <
.05; perceived helpfulness, B = .36, t (69) = 4.87, p < .05. Moreover the combined effect
of these variables predicted accurately the educational attainment of subjects, R = .82; R ’
= .80, F (6, 69) = 23.51, p < .05. (See Table 3 for percentages for gender, parental
income, pregnancy, instrumental messages, loss-flamed messages, helpfulness, and
message repetition by educational attainment) (see Table 4 for means for sibling dropout
by educational attainment).

Table 3

Percentages for Gender, Parental Income, Pregnancy, Instrumental Messages, Loss-

fiamed Messages, Helpfulness, and Message Repetition by Educational Attainment

 

 

 

Educational Attainment
N Dropout, Dropout, Graduate, Graduate,
No GED GED No college College
Gender
Male 32 18% 10% 6% 6%
Female 48 7% 15% 19% 19%
Parental Income
$15,000 or below 13 6% 6% 3% 16%
$16,000-25,000 16 11% 7% 0% 20%
$26,000-35,000 13 5% 6% 3% 16%
$36,000-50,000 l2 1% 2% 11% 15%
$50,000 or above 26 1% 2% 9% 33%
Pregnancy
No 70 18% 23% 21% 25%
Yes 10 6% 3% 4% 0%
Instrumental Messages
No 10 5% 3% 4% 1%
Yes 70 20% 22% 21% 24%
34

Table 3 (cont)

Loss-flamed Messages

 

No 5 5 l 2% l 4% 20% 22%
Yes 25 12% 11% 5% 3%
Helpfulness
NO 3 l 15% l 5% 8% 1%
Yes 49 10% 10% 17% 24%
Message Repetition
Once in awhile 32 9% 10% 10% l 1%
Every month 14 1% 5% 4% 7%
Every week 19 5% 6% 7% 5%
Every day 15 10% 4% 4% 1%
Table 4

Means for Sibling Dropout by Educational Attainment

 

 

 

Educational Attainment
N Dropout, Dropout, Graduate, Graduate,
No GED GED No college College
Sibling Dropout 76 1.6 1.90 .56 .11
( l .70) (1.97) (.86) (.46)

 

Note. Standard deviations in parentheses

Path Model

An examination of the beta weights provides clear evidence that the effect of
perceived helpflrlness is large relative to a number of the other predictors. This outcome
raises the possibility that the effect of some of these predictors may be mediated by
perceived helpfulness. Correlations between perceived helpfulness and a number of

message predictors were examined to explore this possibility. Instrumental support

35

messages and loss-flamed messages were both found to be correlated substantially with
perceived helpfirlness. Thus, a post-hoe causal model was created with perceived
helpfulness mediating the effect of instrumental messages and loss-flamed messages on
educational attainment. The remaining predictors (pregnancy, parental income, gender,
sibling dropout, and message repetition) appeared to exert direct effects on educational

attainment (see Figure 1).

Instrumental 24

 

 

 

 

 

 

 

 

 

.37
Perceived Helpful I —> Educational
Attainment

 

 

 

Loss-Framed

 

 

 

 

 

  
  
 

 

Parent Income

 

 

 

Sibling Dropout

 

 

 

 

Pregnancy

 

 

Gender

 

 

 

Figure 1. Post-hoe path model depicting the relationship between instrumental support,
loss- flaming, perceived helpfulness of message, parent income, gender, sibling dropout,

message repetition and educational attainment.

Table 5 presents the correlations employed to estimate the fit of these model
parameters employing the ordinary least squares criterion (Hunter & Gerbing, 1982). In

the model path coefficients were ample and statistically significant. No errors were found

36

that were greater than what would be expected from sampling error in the population. A

global test for goodness of fit indicated that these data were consistent with the model, x2

(7, 73) = 4.81, p = .68.

Table 5

Correlations for Variables in Path Model

 

 

Variable 1 2 3 4 6 7 8 9
Instrumental Support
Loss Framed .17
Parental Income .20 -.29*
Sibling Dropout .11 .27“ -.30*
Pregnancy -.20 .07 -. 1 7 .06
Gender .00 -.11 .20 -.10 .08
Message Repetition .00 .16 -.07 .02 .00 .05
Perceived Helpfirlness .24“ -.18 .16 -.29 .07 .08 .08
Educational Attainment .14 -.35** .62" -.42** -.24"' -.34*"' .23" .45"

 

Note. N = 80. * p < .05

37

Chapter 4
Discussion

In this study educational attainment was investigated as an outcome of the
number, valence, and type of support message. The results were not consistent with the
first hypothesis concerning differences in the number of messages recalled by subjects
with varying levels of educational attainment. When considering the relationship between
valence of the message and educational attainment only perceived helpfulness was found
to be an important predictor of educational attainment.

Hypotheses 2 and 3 predicted relationships between emotional support and
educational attainment and instrumental support and attainment. Participants with higher
levels of educational attainment did not recall more emotional support type messages
than those with lower educational attainment. Additionally, participants with higher
levels of attainment did not recall more instrumental support type messages than
participants with lower levels of attainment Thus, results were not consistent with these
hypotheses. In sum, only participants’ reports of the helpfulness of the message impacted
their level of educational attainment.

Perceived helpfulness of the message was treated as an alternate method of
measuring message valence. In a study assessing the measurement of enacted social
support in personal relationships, Goldsmith, McDermott, and Alexander (2000) found
that the adjective helpful refers to the informational and instrumental benefits of an
interaction and is associated with information that is viewed as knowledgeable, useflrl,
and generous. The perceived helpfirlness of messages in this study impacted educational

achievement such that participants with higher levels of educational achievement

38

reported receiving more helpful messages than those with lower levels of achievement.
Further analyses were employed to understand what makes a message perceived to be
more or less helpful.

According to the post-hoe causal model, perceived helpfulness mediated the
impact of instrumental messages on educational achievement. Messages are perceived as
more helpflrl if they provide some form of tangible assistance, such as physical assistance
or aid, in the form of advice or guidance. When dealing with issues involving educational
attainment (e. g., grades) students may perceive messages that help them solve a problem
as more helpful than emotional messages that express care, concern, or sympathy. These
results are a departure from what has been found in previous research. Messages
conveying emotional support have regularly been identified by participants as either the
most helpfirl or one of the most helpful forms of assistance they recall receiving fl'om
others (Barbee, Derlega, et al., 1998; Caplan & Samter, 1999; Cramer, 1990; Dakof &
Taylor, 1990; Dunkel-Schetter, 1984; Goldsmith, 1994; Sullivan, 1996). Instrumental
messages may have been viewed as more helpful in this study because the context
surrounding the messages may lend itself to support recipients seeking out advice or
guidance rather than concern or sympathy. Many of the studies that have assessed the
perceived helpfulness of emotional support versus instrumental support have examined
how supportive messages have affected people with serious health problems (e. g.,
Chesler & Barbarin, 1984; Dakof & Taylor, 1990; Dunkel-Schetter, 1984) where
negative consequences cannot necessarily be prevented or altered In this study, subjects
may feel that the negative consequences surrounding educational attainment could be

altered; thus instrumental messages were seen as particularly helpful.

39

Additionally, according to the path model there is evidence that the helpfulness of
a message also mediates the relationship between loss-flamed messages and educational
attainment. Loss-flamed messages emphasize the undesirable consequences associated
with noncompliance (O’Keefe & Jensen, 2005). An example of a message coded as a
loss-flamed appeal is, “If you don’t finish school you will be homeless and poor.”
Noncompliance in this example is not graduating, or not finishing school, and the
undesirable consequence of noncompliance is becoming homeless and poor. The more
subjects recalled loss-flamed messages the less helpful they viewed the message. Loss-
fl'amed messages have traditionally been shown to be significantly more engaging than
their gain-framed counterpMs (e.g., Tversky & Kahneman, 1981) because appeals
emphasizing potential losses will presumably be more engaging (more attention-getting,
more thought provoking, etc.) Yet, according to a meta-analysis by O’Keefe and Jensen
(2005) loss-flamed messages did not generate significantly greater message processing
than gain-flamed messages. In the present study there were no differences in gain-flamed
versus loss-flamed messages recalled by subjects with different levels of attainment, thus
it seems loss-flamed messages were not more engaging, or memorable, and were not
recalled more readily than gain-flamed messages. Conversely, how the messages were
viewed (helpful versus unhelpful) did differ concerning different levels of educational
attainment. Loss-flamed messages may not be more memorable but the specific wording
of such messages may impact how helpful a message is in the specific context of
educational attainment. Specifically, an increase in loss-flamed messages will lead to a

decrease in perceived helpfulness of the message.

40

Along with the mediating effects of helpfulness, the causal path model reported
also includes direct effects of educational attainment. A number of demographic factors
were found to impact educational attainment. Level of parental income, which served as a
measure of socioeconomic status (SES), was a strong predictor of educational attainment.
This result corresponds to previous findings concerning the impact of SES on educational
attainment (Ekstrom, Goertz, Pollack, & Rock, 1986; Fine, 1986; Johnston, Markle, &
Harshbarger, 1986; McDill, Natriello, & Pallas, 1985; Rumberger 1983; Wehlage &
Rutter, 1986). Additionally, there were important differences between males and females
concerning levels of educational attainment. Females reported higher levels of
educational attainment than males, which is consistent with previous research findings on
gender differences in attainment (Battin-Pearson, Newcomb, Abbott, Hill, Catalano, &
Hawkins, 2000; Bickel, 1989; McNeal, 1997; Rumberger, 1987). Moreover, pregnancy
was found to impact educational attainment, which also replicates previous findings
(F orste & Tienda, 1992; Geronimus & Korenman, 1992; Rumberger, 1983). The number
of siblings who have dropped out is another important demographic predictor of
educational attainment, one that has not received as much attention in previous research.
According to the causal model, as the number of subjects’ siblings who dropped out
increased educational attainment decreased. It may be easier, and more acceptable, for
students to drop out of school if they have older brothers or sisters or both who have
made the same choice. Finally, the number of times subjects recalled hearing a message
also impacted their educational attainment. Subjects who recalled hearing the message
more often reported higher levels of educational attainment. This finding provides

important information to communication researchers concerning the impact of

41

memorable messages. Hearing a message one time may not be enough to impact a
decision-making process like deciding to stay in school versus dropping out. Hearing a
message repeatedly may serve to reinforce a decision that is based on the initial
information in the message. It may not be enough to hear a message one time; instead a
message’s impact may increase with subsequent repetitions.

The variables included in the post-hoe path model were not the only variables
assessed concerning educational attainment. Previous research on school success has
revealed a variety of family and school context variables that have been linked to
educational attainment. These context variables were assessed in order to determine if
their effects were replicated in the present study.

Previous research has found a strong relationship between parental level of
education and the educational attainment of their children (Ekstrom, Goertz, Pollack &
Rock, 1986; Fine, 1986; Johnston, Markle & Harshbarger, 1986; McDill, Natriello, &
Pallas, 1985; Rumberger, 1983; Wehlage & Rutter, 1986), yet parental level of education
did not impact subjects’ educational attainment in this study. Parental level of education
may not have replicated because subjects may not have been able to recall accurately the
level of educational achievement for their parent(s). In addition, subjects who did not
graduate from high school had a more difficult time recalling parental level of education
achieved, and perhaps overestimated their parent(s) level of achievement. Parental
aspirations regarding school success has also been linked with educational attainment
(Astone & McLanahan, 1991; Ekstrom et al., 1986; F agan & Pabon, 1990; McCombs &
Forehand, 1989; Rumberger, 1983, Rumberger et al., 1990), but was not predictive of

attainment in this study. Parental aspirations were measured by the following item: “Was

42

your performance in school important to your parent(s)?” Almost 90% of participants
reported that their performance was important. This item may have been too vague and
may not have effectively measured parental aspirations toward educational achievement.
Future research would benefit fi'om incorporating multiple indicators of parental
aspirations in order to further understand the relationship between aspirations and
educational attainment.

School context variables have also been linked with educational attainment. Low
academic achievement, typically assessed with GPA, has been the primary aspect of
school performance studied in the last 20 years, and has consistently been one of the
strongest predictors of dropping out of school (Cairns, Cairns, & Neckerman, 1989;
Ekstrom, Goertz, Pollack & Rock, 1986; Fagan & Pabon, 1990; Rumberger, 1983).
Subject reports of their GPA in high school were not predictive of educational attainment,
although subject reports may be unreliable due to memory decay or social desirability
issues (e.g., subjects who did not do well in school may have over reported their GPA
while in high school). Previous research has found that students who worked more than
15 hours per week while in high school have also been found to be more likely than their
peers to drop out of high school (Battin-Pearson, Newcomb, Abbott, Hill, Catalano, &
Hawkins, 2000; Bickel, 1989; Fine, 1986; McDill et al., 1985, McNeal, 1997;
Rumberger, 1987; Wehlage & Rutter, 1986). Yet working students were not more likely
to have lower educational attainment than students who did not work while in high
school. In the present study subjects reported working an average of 13.15 hours per
week (SD = 12.04), which may not have been enough time to cause substantially

decreased focus on educational attainment. Furthermore, dropouts have reported taking

43

part in fewer after-school activities than their graduating counterparts (Ekstrom et al,
1986; Johnston et al., 1986; Mahoney & Cairns, 1997; Wehlage & Rutter, 1986), yet no
differences were found on educational attainment between students who did and did not
participate in extracurricular activities. It is important to note that subjects who reported
involvement spent an average of 6 hours per week participating, which is a very low level
of involvement. Perhaps a sample with higher participation would yield different results
concerning attainment. Finally, grade retention has been found to be an important and
consistent predictor of high school dropout (Alexander, Entwistle, & Horsey, 1997;
Roderick, 1994). In this study grade retention was significantly correlated with
educational attainment, r = -.48, p < .05, yet this effect was reduced substantially when
regressed on attainment along with the parental income, gender, pregnancy, sibling
dropout, message repetition, and message helpfulness. Clearly, its effects arise largely
because it is correlated substantially with important antecedents of educational
attainment.
Limitations

One limitation of this study is that subjects predominantly recalled instrumental
support messages. Very few emotional support messages were actually recalled. The
predominance of instrumental messages might have been due to how message recall was
solicited, namely subjects were asked to share verbally their messages with the
experimenter. There may have been an increase in the number and type of messages
recalled if subjects had been asked to write out their messages. A pilot study was
conducted in which subjects were asked to write down all the messages they remembered

receiving about educational attainment. On average, subjects recalled three messages

using this method. Another explanation for the predominance of instrumental messages is
they simply may be easier to recall. Future research examining instrumental messages
may benefit fl'om incorporating both verbal and written message recall to address this
issue.

A second limitation of this study is that all variables were measured with one
item. The interview protocol asked about a wide variety of topics but as it was important
to limit the length of the interview to avoid subject fatigue, thus single items were
employed. Although the reliability of one-item measures could not be determined the
questions exhibited face validity.

A third limitation is how the sample was collected. Participants were recruited
through a snowball sampling technique, which is a convenient, non-random method of
obtaining subjects. Generalizations based on results garnered through this sampling
method should be limited as the sample in this study may not accurately reflect what is
occurring in the general population. Future research may benefit from incorporating a
random sample of subjects when assessing issues associated with educational attainment.

A final limitation is the hypotheses posited in this study lacked a unifying
theoretical structure for predicting dropout. Future research must strive to incorporate an
a priori theoretical explanation for the influence of such variables to allow for the
systematic study of family communication variables. Uncertainty reduction theory
(Berger & Calabrese, 1975) may be an important flamework for explaining what types of
supportive interactions occur and how they affect educational attainment. The
opportunity to talk over feelings and reactions with others is a way to clarify and reduce

one’s uncertainty toward stressful communication (Albrecht & Adelman, 1984). Berger

45

and Calabrese (1975) claim that people are motivated to communicate with one another
in order to describe, predict, and explain another’s behavior. Perhaps parents attempt to
reduce uncertainty by talking to their child about school and school success. Parents may
use instrumental support messages to achieve this goal as informational messages may
provide more uncertainty reduction than emotional support messages because they
provide advice about how to address negative circumstances that can be altered.

Future Directions

This study is an initial attempt at understanding how different aspects of messages
in the home may affect educational attainment. The post-hoe causal path model presented
provides information about what variables may impact student achievement in school and
beyond. In order to determine the validity of the model, it is recommended that a
replication with formal hypotheses about the model be conducted. Additionally, a
longitudinal study assessing the impact of parental communication and educational
attainment is suggested in order to understand more fully the relationship between these
constructs.

The long-term utility of this analysis is to use the information garnered to develop
effective school dropout prevention messages. Communication in the home is not the
only important social process that can be addressed in research on dropouts. Messages
shared in the school context as well as messages exchanged between peers may also be
important areas of study for future research on educational attainment. Exploration of the
differences between varying relationships associated with school success is needed to

identify effective methods of dropout prevention and remediation (Lecompte & Goebel,

46

1987). Understanding how the process of communication affects dropout may provide

valuable information for new prevention efforts.

47

Footnotes

‘ Conventionally race was thought to be the most important predictor of school
dropout, but research has demonstrated that when socioeconomic status is held constant,
race no longer matters (Rumberger, 1995).

2 There are differing categorizations of support types. Discrepancies often arise
when differentiating informational (providing advice and information) and instrumental
support (which includes offering tangible assistance or physical assistance in the form of
advice). Some studies make the distinction between emotional and instrumental support
as the two major areas of support types (e.g., Kaufinann & Beehr, 1986), while others
studies make the distinction between emotional and informational support as the two
major areas of support types (Sullivan, 1996). This study uses the former distinction of

emotional and instrumental types as the two main categorizations of support.

48

APPENDICES

49

APPENDIX A
Consent Form

Dropping Out of School:
Assessing the Effect of Supportive Messages from Family about Staying in School

Family functioning and school achievement are linked in research on dropouts, yet
familial communication factors are often overlooked in research concerning a student’s
decision to stay in, or drop out of school. The purpose of this study is to examine
messages people recall having received while in high school in order to understand if
there is a relationship between these messages and people’s level of educational
attainment. It is believed that the project will assist in a better understanding of what
types of messages can be used in campaigns targeted at school success.

As a part of this research, you are being asked to participate in an interview where you
will be asked a series of questions about messages that you recall from high school. You
will also be asked to complete a brief survey. Participation in this study will take
approximately 30 minutes. Interviews will be conducted in 455 Corn Arts on the campus
of Michigan State University. Foreseeable risks involved with participation in this study
include loss of privacy, and participation may cause feelings of embarrassment and
awkwardness. Foreseeable benefits of participation are none. Participation is voluntary,
you may choose not to participate at all, or you may refuse to participate in certain
procedures or answer certain questions or discontinue your participation at any time
without penalty or loss of benefits. You will be paid $15 for your participation in this
study, regardless of whether you actually complete the entirety of the interview. All
interviews will be tape recorded, but no identifying information will be associated with
any of the tapes. The primary researcher will be the only person with access to the tapes
and all tapes will be stored in a locked filing cabinet for 1 year. Your privacy will be
protected to the maximum extent allowable by law. The following precautions will be
taken to protect your confidentiality. No individual names or other identifying
information will be used in any reports or publications that may result from this study.

Your participation in the study would be greatly appreciated. However, please know that
you may refuse to participate without any penalty. If you have any questions about the
study, please contact the primary investigator (Frank Boster, 567 Corn Arts and Sciences,
517-355-1514, boster@msu.edu) If case you have questions or concerns about your
rights as a research participant, or are dissatisfied at any time with any aspect of this
study, you may contact - anonymously, if you wish - Peter Vasilenko, Ph.D., Chair of the
Human Research Protection Program by phone: (517) 355-2180, fax: (517) 432-4503,
email address: irb@msu.edu, or regular mail: 202 Olds Hall, East Lansing, MI 48824.

If you agree to participate in this study, please write your name and sign your name on
the lines below.

Printed Name:

 

50

Signature:

 

By initialing below you are indicating your permission to be tape-recorded:

Initials

 

Thank you for your time and cooperation.
Sincerely,

Renee Strom, MA.

51

APPENDIX B
Interview Protocol

Overall Context While in School

1. How many years did you attend high school?

2. Did you get your high school diploma?

2a. (If did get diploma) At any time did you drop out while working towards your high
school diploma?

2b. (If didn ’t get diploma) Have you worked on getting a GED (General Education
Diploma)?

2c. (If working towards GED) How long have you worked towards getting your GED?

2d. (Ifdidn ’t get diploma) Did you obtain your GED?

3. Have you attended college?

3a. (If yes) For how long?

3b. (If yes) Did you obtain a college degree?

3c. (If yes) What type of degree? (2 year, 4 year, Masters, etc.)

4. Now I would like you to think about when you were in high school and try to
remember as many aspects of high school that you can (e.g., school activities, school
success, grade point average, etc.) Think specifically about a period of time when you
may have been struggling in school and/or a time when you were doing exceptionally
well in high school. Can you think of a time like this?

4a. How old were you during this time?

4b. What year was this?

4c. What time in the school year was it?

5. Tell me about any events (at home or at school, with fliends, etc.) that occurred during

this time flame.

5a. (Probe) Did any of these events affect (positively or negatively) how you were doing

in school?

5b. (Probe if events did aflect them) How so?

6. (For those who dropped out) Why did you decide to drop out of high school?

7. Were you ever held back a grade in school?

School Context
1. Where did you go to high school?
1a. (Probe) Did you attend high school anywhere else?
2. What size of school did you attend (how many people in your graduating class)?
3. Did you like school, dislike school, both like or dislike, or neither like nor dislike
school while attending it?
3a. (If liked school) What things did you especially like about school?
3b. (Probe if necessary) Anything else?
3c. (If disliked school) What things did you especially dislike about school?-
3d. (Probe if necessary) Anything else?
4. Looking back on school do you feel differently about it now than when you were
attending?

52

4a. (If subject feels diflerently) Would you say you feel more positive about school, more
negative about school, or both?
5. Approximately what was your grade point average while in high school?
5a. (Probe) How well do you think you did academically?
6. Did you have a job while attending high school?
6a. (If yes) What did you do?
6b. (If had a job) About how many hours a week did you work?
7. Were you involved in extracurricular activities in high school? (band, sports, yearbook,
etc.)
7a. (If yes) What kind of activities?
7b. (If yes) How long did that take each week?
8. Were there particular teachers in high school that you liked a lot?
8a. (If yes) Tell me about them.
9. Were there particular teachers in high school you disliked a lot?
9a. (If yes) Tell me about them.
Family Context

1. While in high school did you live with both parents?

1a. (If no) Did you live with one parent?

lb. (If lived with one parent) Which parent did you live with?

1c. (If didn ’t live with either parent) What was your living situation?

2. Do you have any brothers and sisters?

2a. (If yes) How many brothers and/or sisters?

2b. (Probe) Are they biological, step, half or adopted siblings?

2c. (if they have siblings) What are the age differences between you and your sibling(s)?

2d. (If they have siblings) While in high school did you confide in, or tell important
information, to any of your siblings?

2e. (Probe if yes) Tell me about that.

2f. (If they have siblings) How many of your sibling(s) did well in school?

2g. (If they have siblings) How many of your sibling(s) dropped out of high school?

2h. Did your sibling(s) go on to college?

3. Did your parent(s)/primary caregiver graduate flom high school?

4. Did your parent(s)/primary caregiver go on to college?

4a. (If yes) Did they get a college degree?

4b. (If yes) What type of degree (2 year, 4 year, Masters, Phd)?

5. Did either or both of your parent(s)/primary caregiver work?

5a. (If yes) Where did your parent(s)/primary caregiver work while you were in high

school?

5b. (Probe) How many hours a week did they work?

Sc. (If parent/primary caregiver worked) Did they like their job a lot or not much?

6. Was your performance in school important to your parent(s)/primary caregiver?

6a. (If yes) In what way did your parent(s)/primary caregiver show you that your

performance was important?

6b. (If no) In what way did your parent(s)/primary caregiver show you that your

performance was unimportant?

53

7. Were there other things your parent(s)/primary caregiver did that helped you achieve
'what you needed to in school?
7a. (Probe if necessary) Like what?
8. Were there other things your parent(s)/primary caregiver didn’t do that helped you
achieve what you needed to in school?
8a. (Probe if necessary) Like what?
Memorable Messages

Now think about the things people said to you while you were in high school. These
things could be positive or negative in nature. Specifically think about the things people
said about education and being successful in school and think about what effect the things
they said had on you.

When you think about when you were in high school what do you remember your
parent(s)/primary caregiver telling you about education? Please tell me the things you
remember hearing.

*Addressing the first message

1a. Who said this to you?

1b. Did you ask for this information?

1c. (Probe if necessary) Did your mom/dad/primary caregiver tell you this a lot?

1d. (If they only talk about mom) Did your dad ever say anything like this to you?

1e . Did anybody else ever say this to you as well?

If. (If no) Did siblings, peers, teachers ever say this to you? ,

lg. The first time someone said this to you did you take it as a positive message about
education, a negative message about education, or as a neither positive or negative
message about education?

1h. (If positive) How positive? (somewhat positive, very positive?)

1i. (If negative) How negative? (somewhat negative, very negative?)

1 j. (Probe) Did this information discourage you from persuing school or encourage you
to try harder?

1k. (Probe) Why?

11. (Probe if necessary) How often did people say this kind of thing to you (every day, a

few times, once)?

1m. (If they heard it ofien) After you heard people say this a lot of times did you still

view it in the same way as you did the first time you heard it? (Because sometimes when

people say things over and over the meaning changes for us)

1n. (Probe if they didn ’t see it the same way after hearing it often) What changed?

10. Did what they said to you come in response to something you said?

1p. Was what they said to you helpflrl?

lq. (If yes) How so?

1r. (If no) Why not?

Is. When they said this to you did you respond?

It. (If yes) What did you say?

lu. (Probe) How positive/negative was your response?

54

1v. Did the person who told you this ever help you with your homework?

1w. (If yes) How often did they help you with your homework (how many nights during
an average week)?

1x. (If no) Did they ever talk to you about your homework (offer to help, monitor, check
answers?)

1y. Did your parent(s)/primary caregiver often go to parent-teacher conferences?

12. (If yes) What percentage of conferences did your parent(s)/primary caregiver attend?
*Addressing the second message

2. What else do you remember your parent(s)/primary caregiver telling you about

education?

2a. Who said this to you?

2b. Did you ask for this information?

20. (Probe if necessary) Did your mom/dad/primary caregiver tell you this a lot?

2d. (If they only talk about mom) Did your dad ever say anything like this to you?

2e . Did anybody else ever say this to you as well?

2f. (if no) Did siblings, peers, teachers ever say this to you?

2g. The first time someone said this to you did you take it as a positive message about
education, a negative message about education, or as a neither positive or negative
message about education?

2h. (If positive) How positive?

2i. (If negative) How negative

23'. (Probe) Did this information discourage you from persuing school or encourage you
to try harder?

2k. (Probe) Why?

21. (Probe if necessary) How often did people say this kind of thing to you (every day, a
few times, once)?

2m. (If they heard it often) After you heard people say this a lot of times did you still
view it in the same way as you did the first time you heard it?

2n. (Probe if they didn’t see it the same way after hearing it often) What changed?

20. Did what they said to you come in response to something you said?

2p. Was what they said to you helpful?

2q. (If yes) How so?

2r. (If no) Why not?

2s. When they said this to you did you respond?

2t. (If yes) What did you say?

2u. (probe) How positive/negative was your response?

2v. Did the person who told you this ever help you with your homework?

2w. (If yes) How often did they help you (how many nights during an average week)?

2x. (If no) Did they ever talk to you about your homework (offer to help, monitor, check
answers?)

2y. Did your parent(s)/primary caregiver often go to parent-teacher conferences?

22. (If yes) What percentage of conferences did your parent(s)/primary caregiver attend?

* (Repeat these steps until subject can’t recall anymore messages)

55

Peer Context

1. Did you have fliends in high school who dropped out of high school?

la. (If yes) How many dropped out?

1b. (If yes) How did that affect you?

1c. (If yes) Did you have other fliends who didn’t drop out of high school?

1d. (if yes) Were they involved in school?

1e. (if fiiends didn’t drop out) Were your fiiends involved in school?

If. (If they had fiiends) As far as you know how many of your fliends ever considered
dropping out of high school?

1h. (If they had friends) How many of your fiiends did well in high school?

1i. (if they had fiiends) Did your friends ever skip school?

1 j. (if yes) How often?

1k. (if yes) Did you ever skip school with your friends?

1k. (If they had friends) Did any of your fliends go to college?

56

APPENDIX C
Demographic Questions

1. Please circle the average amount of money you currently make in a year.

$15,000 $16,000- $26,000- $36,000- $50,000
or Below $25,000 $35,000 $50,000 or Above

2. Please circle the average amount of money your primary caregiver(s) made a year
while living with them.

$15,000 $16,000- $26,000- $36,000- $50,000
or Below $25,000 $35,000 ‘ $50,000 or Above

3. Please circle the highest educational level your father has achieved.

Some High School Some 2 year College Bachelor Masters Doctoral
High School Degree College Degree Degree Degree Degree

Other:

 

4. Please circle the highest educational level your mother has achieved.

Some High School Some 2 year College Bachelor Masters Doctoral
High School Degree College Degree Degree Degree Degree

Other:

 

5. Please circle the highest educational level achieved by the person you recalled above
(if not your mother or father).

Some High School Some 2 year College Bachelor Masters Doctoral
High School Degree College Degree Degree Degree Degree

Other:

 

6. Please circle the highest educational level you have achieved thus far:

Some High School Some 2 year College Bachelor Masters Doctoral
High School Degree College Degree Degree Degree Degree

Other:

 

7. What is your gender? Male Female

57

8. How old are you? years old

9. What is your ethnicity? (Please circle the number next to the race/ethnicity that best
describes you)
1) Asian/Asian-American
2) Black/African-American
3) Hispanic, Latina
4) Pacific Islander
5) Native American/American Indian
6) White, European American
7) Multicultural Mixed Race
8) Other

Thank you for your participation!

58

APPENDIX D
Instructions for Coders and Coding Scheme for Message Types

Instructions for Coders:

Definition of Emotional Support: expressions of care, concern and sympathy,
understanding, and encouragement, and indicates acceptance of a person.

KEY WORDS:

Acceptance for the target

Love for the target

Positive regard for the target

Affection for the target

Expresses concern for the situation

Expresses care and interest about the target’s current situation
Expresses availability to the target

Expresses alliance with the target

Definition of Instrumental support: rendering tangible assistance, such as physical
assistance or aid in the form of advice or guidance.

KEY WORDS:

Help in defining, understanding, and/or coping with problematic events
Offers of information about resources

Suggest alternative courses of action

Provide advice about what to do

Help in obtaining needed services

Provide a new way of seeing the problem

Referral to some other helper

Assisting in active coping efforts

Providing financial assistance or material resources

59

Support Message types

 

Subject #

Message #

Type message here

Support Code
Emotional = 1
Instrumental = 2
Both Emotional &
Instrumental = 3
Neither Emotional
or instrumental = 4

 

N/A 1

Message 1

My mom always said she was
really proud of how I was doing
in school

1

 

N/A 2

Message 1

My parents told me that it was
important that I get my high
school diploma or I was going
to struggle in life.

 

Message 2

My mom said she would buy
me a car if I graduated from
high school.

 

N/A 3

Message 1

My parents told me that they
were there for me no matter
what happened at school. They
said, “We will help you get your
grades up. The best way for you
to do that is to come to us with
your questions and problems.”

 

N/A 4

Message 1

My mom told me I should finish
school but those were just
words. Her actions never
supported her words because
she never saw to it that I

actually did it.

 

Practice

Message 1

My dad always told me was that
he didn’t want me to go down
the same path as he did and not

graduate.

 

Practice

Message 1

I remember hearing that school
is really important.

 

Practice

Message 1

My mom just always told me
how I am a screw-up.

 

Practice

 

 

Message 1

 

My mom and dad would talk to
me about how they were
concerned about how I was
doing in school. They wanted to
make sure I was okay and that I
wasn’t going to flunk out.

 

 

60

 

 

 

Practice

 

Message 1

 

My mom would ask me what
my plans were for after high
school. She would say all I have
to worry about is school right
now.

 

61

 

 

APPENDIX E
Instructions for Coders and Coding Scheme for Kernel States and Framing

o Kernel States
0 Basic, root state mentioned in the message’s description of the
consequence under discussion

 

O Code as
- 1) desirable kernel state (e.g., graduating from high school, getting high
school diploma, going to college, finishing school, importance of school)

- 2) undesirable kernel state (not graduating from HS, not getting high
school diploma, dropping out, not finishing school, unimportance of school)

- 3) a combination of desirable and undesirable kernel states (mentions
graduating from high school, or getting high school diploma, or going to
college, or finishing school, or importance of school AND not
graduating from HS, or not getting high school diploma, or dropping out,
or not finishing school, or unimportance of school)

- 4) indeterminate due to phrasing (doesn’t mention any of the above)

Ex: “You will be homeless if you drop out of high school”
- This message has an undesirable kernel state (dropping out)
- undesirable consequences of noncompliance by discussing an
undesirable kernel state that will happen

Ex: “If you graduate from high school you can get your own place”
- describes a desirable kernel state (graduating from HS) that will be
attained by compliance

° Gain vs. loss framirrg

o 1) Gain-framed: The appeal emphasizes the desirable consequences
associated with compliance with the advocated viewpoint
I Ex: “If you graduate from high school you can get your own place’
- This message emphasizes the desirable consequences of
compliance.

9

o Loss-framed-undesirable consequences associated with noncompliance
0 Ex: “You will be homeless if you drop out of high school”
- This message discuss the undesirable consequences of
noncompliance

62

 

Subject

Message

Type message here

Kernel State
l=undesirable
2=desirable

3= both

4= indeterminate

Framing
1=ncgative
2=positive

3= both
4=indeterminate

 

N/A 1

Message 1

My mom always said she was
really proud of how I was
doing in school

4

4

 

N/A 2

Message 1

My parents told me that it was
important that I get my high
school diploma or I was going
to struggle in life.

 

Message 2

My mom said she would buy
me a car if I graduated from
high school.

 

N/A 3

Message 1

My parents told me if I quit
school I would never amount

to anything.

 

N/A 4

Message 1

My mom told me I should
finish school but those were
just words. Her actions never
supported her words because
she never saw to it that I
actually did it.

 

N/A 5

Message 1

If you don’t take graduating
seriously, you’re never going
to amount to anything.

 

N/A 6

Message 1

School is important. It’s
important that I go. She told
me my education should be my
first priority.

 

Practice

Message 1

My dad always told me was
that he didn’t want me to go
down the same path and
struggle as he did and not
graduate.

 

Practice

Message 1

I remember hearing that school
is really important.

 

Practice

Message 1

My mom just always told me
how I am a screw-up.

 

 

Practice

 

Message 1

 

My mom and dad would talk
to me about how they were
concerned about how I was
doing in school. They wanted
to make sure I was going to
graduate and get a job.

 

 

 

63

 

 

 

 

Practice

 

Message 1

 

My mom would ask me what
my plans were for after high
school. She would say all I
have to worry about is school
right now.

 

64

 

 

 

REFERENCES

65

REFERENCES

Abbey, A., Abramis, D. J ., & Caplan, R. D. (1985). Effects of different sources of social
support and social conflict on emotional well-being. Basic and Applied Social
Psychology, 6, 111-129.

Alexander, K. L., Entwistle, D. R., & Horsey, C. S. (1997). From first grade forward:
Early foundations of high school dropout. Sociology of Education, 70, 87-107.

Alexander, K. L., Entwistle, D. R., & Kabbani, N. S. (2001). The dropout process in life
course perspective: Early risk factors at home and school. Teachers College
Record, 103(5), 760-822.

Andrews, G., Tennant, C., Hewson, D., & Schonell, M. (1978). The relation of social
factors to physical and psychiatric illness. American Journal of Epidemiology,
108, 27-35.

Antonucci, T. C., & Jackson, J. S. (1987). Social support, interpersonal efficacy, and
health: A life course perspective. In L. L. Carstensen & B. A. Edelstein (Eds),
Handbook of clinical gerontology (pp. 291 -31 1). New York: Pergamon Press.

Astone, N. M., & McLanahan, S. S. (1991). Family structure, parental practices and high
school completion. American Sociological Review, 56, 309-320.

Baker, D. P., & Stevenson, D. L. (1986). Mother’s strategies for children’s school
achievement: Managing the transition to high school. Sociology of Education, 59,
156-166.

Barbee, A. P., Derlega, V. J ., Sherbume, S. P., & Grimshaw, A. (1998). Helpful and
unhelpful forms of social support for HIV-positive individuals. In V. J. Derlega &
A. P. Barbee (Eds), HIV and social interaction (pp. 83-105). Thousand Oaks,
CA: Sage.

Barrera, M. (1988). Social support in the adjustment of pregnant adolescents. In A. Vaux
(Ed.), Social support: Theory, research, and intervention (pp. 69-96). New York:
Praeger.

Barro, S., & Kolstad, A. (1987). Who drops out of high school? Findings from high
school and beyond. (Report No. CS 87-3970). Washington, DO: U. S.
Department of Education, National Center for Education Statistics.

Barton, W. H., Watkins, M., & J arjoura, R. (1997). Youths and communities: Toward
comprehensive strategies for youth development. Social Work, 42, 483-493.

66

Battin-Pearson, S., Newcomb, M. D., Abbott, R. D., Hill, K. G., Catalaon, R. F., &
Hawkins, D. J. (2000). Predictors of early high school dropout: A test of five
theories. Journal of Educational Psychology, 92, 568-582.

Beauvais, F ., Chavez, E. L., Oetting, E. R., Deffenbacher, J. L., & Cornell, G. R. (1996).
Drug use, violence, and victimization among White Americans, Mexican
Americans, and American Indian dropouts, students with academic problems, and
students in good standing. Journal of Counseling Psychology, 43, 292-299.

Beehr, T. A. (1985). The role of social support in coping with organizational stress. In T.
Beehr & Bhagat, R.(Eds.), Human stress and cognition in organizations: An
integrated perspective (pgs. 375-398). New York: Wiley.

Berger, C., & Calabrese, R. (1975). Some explorations in initial interactions and beyond:
Toward a developmental theory of interpersonal communication. Human
Communication Research, 1, 99-112.

Bickel, R. (1989). Post-high school opportunities and high school completion rates in an

Appalachian state: A near replication of Florida research. Youth and Society, 21,
61-84.

Billings, A. G., & Moos, R. H. (1982). Social support and firnctioning among community
and clinical groups: A panel model. Journal of Behavioral Medicine, 5, 295-312.

Bryk, A. S., & Thum, Y. M. (1989). The effect of high school organizations on dropping
out: An exploratory investigation. American Educational Research Journal, 26,
353-3 83.

Bureau of Justice Statistics (1995). Sourcebook of criminal justice statistics.

Burleson, B. R. (2003). Emotional support skills. In J. O. Greene and B. R. Burleson
(Eds.), Handbook of communication and social interaction skills (pp. 551-594).
Mahwah, NJ: Lawrence Erlbaum Associates.

Burleson, B. R. (1994b). Comforting messages and features, functions, and outcomes. In
J. A. Daly and J. M. Wiemann (Eds.), Strategic interpersonal communication (pp.
135-161).

Burleson, B. R. & MacGeorge, E. L. (2002). Supportive communication. In M. L. Knapp
& J .A. Daly (Eds.), Handbook of interpersonal communication, 3rd edition (pp.
374-424). London: Sage.

Burleson, B. R., & Samter, W. (1985a). Consistencies in theoretical and naive
evaluations of comforting messages. Communication Monographs, 52, 103-123.

67

Cairns, R. B., Cairns, B. D., & Neckerman, H. J. (1989). Early school dropout:
Configurations and Determinants. Child Development, 60, 1437-1452.

Caplan, G. (1976). The family as a support system. In G. Caplan & m. Killilea (Eds.),
Support systems and mutual help (pp. 19-36).

Caplan, S. E., & Samter, W. (1999). The role of facework in younger and older adults’
evaluations of social support messages. Communication Quarterly, 47, 245-264.

Census Bureau (1999). Statistical abstract of the United States. Washington DC: US
Department of Commerce.

Chesler, M. A., & Barbarin, O. A. (1984). Difficulties of providing help in a crisis:
Relationship between parents of children with cancer and their friends. Journal of
Social Issues, 40(4), 113-134.

Chisholm, R. F ., Kasl, S. F., & Mueller, L. (1986). The effects of social support on
nuclear worker responses to the three mile island accident. Journal of
Occupational Behavior, 7, 179-193.

Christenson, S. L., Sinclair, M. F., Lehr, C. A., & Godber, Y. (2001). Promoting
successful school completion: Critical conceptual and methodological guidelines.
School Psychology Quarterly, 26, 468-484.

Clark, S. L., & Stephens, M. A. P. (1996). Stroke patients’ well-being as a function of
caregiving spouses’ helpful and unhelpful actions. Personal Relationships, 3, 171-
184.

Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine,
38, 300-314.

Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and
Psychological Measurement, 20, pp. 37 46.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis.
Psychological Bulletin, 98, 310-357.

Coyne, J. C., Wortrnan, C. B., & Lehman, D. R (1988). The other side of support:
Emotional overinvolvement and miscarried helping. In B. H. Gottlieb (Ed.),
Marshalling social support (pg. 305-330). Newbury Park, CA: Sage.

Cramer, D. (1990). Helpful actions of close fliends to personal problems and distress.
British Journal of Guidance and Counseling, 18, 280-292.

Cutrona, C. E. (1990). Stress and social support: In search of optimal matching. Journal
of Social and Clinical Psychology, 9, 3-14.

68

Cutrona, C. 13., Cole, V., Colangelo, N., Assouline, S. G., and Russell, D. W. (1994).
Perceived parental social support and academic achievement: An attachment
theory perspective. Journal of Personality and Social Psychology, 66 (2), 369-
378.

Cutrona, C. E., & Suhr, J. A. (1994). Social support communication in the context of
marriage: An analysis of couples’ supportive interactions. In B. R. Burleson, T. L.
Albrecht, and I. G. Sarason (Eds.), Communication of social support: Messages,
interactions, relationships, and community (pp. 113-135).

Cutrona, C. E., Suhr, J. A., & McFarlane, R. (1990). Interpersonal transactions and the
psychological sense of support. In S. Duck (Ed.), Personal relationships and
social support (pp. 30-45) London: Sage.

Dakof, G. A., & Taylor, S. E. (1990). Victims’ perceptions of social support: What is
helpful for whom? Journal of Personality and Social Psychology, 58, 80-89.

Dombusch, S. M., Ritter, P. L., Leiderman, P. H., Roberts, D. F. & Fraleigh, M. J.
(1987). The relation of parenting style to adolescent school performance. Child
Development, 58, 1244-1257.

Dryfoos, J. G. (1990). Adolescents at risk: Prevalence and prevention (pgs. 90-91). New
York: Oxford University Press.

Dunkel-Schetter, C. (1984). Social support and cancer: Findings based on patient
interviews and their implications. Journal of Social Issues, 40, 77-98.

Dunkel-Schetter, C., Blasband, D., Feinstein, L., & Herbert, T. (1992). Elements of

supportive interaction: When are attempts to help effective? In S. Spacapron & S.
Oskemp (Eds.), Helping and being helped: Naturalistic studies (pg. 83-114).
Newbury Park, CA: Sage Publications.

Dynarski, M., & Gleason, P. (2002). How can we help? What we have learned form
recent federal dropout prevention evaluations. Journal of Education for Students
Placed At Risk, 7, 43-69.

Ekstrom, R. B., Goertz, M. E., Pollack, J. E., & Rock, D. A. (1986). Who drops out of
high school and why? Findings from a national study. Teachers College Record,
87, 356-373.

Eckenrode, J. (1984). Impact of chronic and acute stressors on daily reports of mood.
Journal of Personality and Social Psychology, 46, 907-918.

69

Ensminger, M. E., & Slusarcick, A. L. (1992). Paths to high school graduation or
dropout: A longitudinal study of a first grade cohort. Sociology of Education, 65,
95-113.

F agan, J ., Pabon, E. (1990). Contributions of delinquency and substance abuse to school
dropout among inner-city youths. Youth and Society, 21, 306-354.

F enlason, K. J ., & Beehr, T. A. (1994). Social support and occupational stress: Effects of
talking to others. Journal of Organizational Behavior, 15, 157-175.

Fine, M. (1986). Why urban adolescents drop into and out of public high school.
Teachers College Record, 8 7, 393-409.

Fitzpatrick, K. M., & Yoels, W. C. (1992). Policy, school structure, and
sociodemographic effects on statewide high school dropout rates. Sociology of
Education, 65, 76-93.

F orste, R., & Tienda, M. (1992). Race and ethnic variation in the schooling consequences
of female adolescent sexual activity. Social Science Quarterly, 73, 12-30.

Franklin, C. (1992). Family and individual patterns in a group of middle-class dropout
youths. Social Work, 37, 338-344.

Geronimus, A. T., & Korenman, S. (1992). The socioeconomic consequences of teen
childbearing reconsidered. Social Science Quarterly, 73, 12-30.

Goldsmith, D. J. (1994). The role of facework in supportive communication. In B. R.
Burleson, T. L. Albrecht, & I. G. Sarason (Eds.), Communication of social
support: Messages, interactions, relationships and community (pp. 29-49).
Thousand Oaks, CA: Sage.

Goldsrrrith, D. J ., McDermott, V. M., & Alexander, S. C. (2000). Helpful, supportive, and
sensitive: Measuring the evaluation of enacted social support in personal
relationships. Journal of Social and Personal Relationships, 1 7 (3), 369-391.

Hammack, F. M. (1987). Large school systems’ dropout reports: An analysis of
definitions, procedures, and findings. In G. Natriello (Ed.), School dropouts:
Patterns and policies (pp. 20-25). New York: Teachers College Press.

Hawkins, J. D., Catalano, R. F ., & Miller, J. Y. (1992). Risk and protective factors for
alcohol and other drug problems in adolescence and early adulthood: Implications
for substance abuse prevention. Psychological Bulletin, 112, 64-105.

Hess, R. S., & Copeland, E. P. (2001). Students’ stress, coping strategies, and school

completion: A longitudinal perspective. School Psychology Quarterly, 16(4), 389-
405. '

70

House, J. S. (1981). Work stress and social support. Reading, MA: Addison-Wesley.

Howell, F. M., & Frese, W. (1982). Early transitions into adult roles: Some antecedents
and outcomes. American Educational Research Journal, 19, 51-73.

Hunter, J ., & Gerbing, D. (1982). Unidimensional measurement, second order factor
analysis and causal models. Research in Organizational Behavior, 4, 267-320.

J anosz, M., LeBlanc, M., Boulerice, B., & Tremblay, R. E. (1997). Disentangling the
weight of school dropout predictors: A test on two longitudinal samples. Journal
of Youth and Adolescence, 26(6), 733-762.

J imerson, S., Egeland, B., Sroufe, L. A., & Carlson, B. (2000). A prospective longitudinal
study of high school dropouts examining multiple predictors across development.
Journal of School Psychology, 38(6), 525-549.

Johnston, J. H., Markle, G. C., & Harshbarger, M. (1986). What research says about
dropouts. Middle School Journal, I 7, 8-11.

Jordan, W., Lara, J ., & McPartland, J. (1996). Exploring the causes of early dropout
among race-ethnic and gender groups. Youth and Society, 28, 62-94.

Kaufrnann, G. M., & Beehr, T. A. (1986). Interactions between job stressors and social
support: Some counterintuitive results. Journal of Applied Psychology, 71, 522-
526. '

Kessler, R. C., & McLeod, J. D. (1985). Sex differences in vulnerability to undesirable
life events. American Sociological Review, 49, 620-631.

Kortering, L. J ., Hess, R. S., & Braziel, P. M. (1997). School drop out. In G. G. Bear, K
M. Minke, & A. Thomas (Eds.), Children ’s needs 11: Development, problems, and
alternatives (pp. 511-521). Betheseda, MD: National Association of School
Psychologists.

Krause, N. (1990). Stress, support, and well-being in later life: Focusing on salient social
roles. In M. A. P. Stephens, J. H. Crowther, S. E. Hobfoll, & D. L. Tennebaum
(Eds.), Stress and coping in later-life families (pp. 122-139). New York:

Hemisphere.

Landis, J .R., & KOch, CG. (1977). The measurement of observer agreement for
categorical data. Biometrics, 33, pp. 159 174.

LaRocco, J. M., House, J. S., & French, Jr., J. (1980). Social support, occupational stress,
and health. Journal of Health and Social Behavior, 21 , 202-218.

71

LeCompte, S. 0., & Goebel, S. D. (1987). Can bad data produce good program planning:
An analysis of record keeping on high school dropouts. Education and Urban
Society, 19, 250-268.

Lehman, D.R., & Hemphill, K. J. (1990). Recipients’ perceptions of support attempts and
attributions for support attempts that fail. Journal of Social and Personal
Relationships, 7, 563-574.

Luszki, M. B., & Schmuck, R. (1963). Pupil perceptions of parental attitudes toward
school. Mental Hygiene, 4 7, 289-299.

Maccoby, E., & Martin, J. (1983). Socialization in the context of the family: Parent-child
interaction. In P. Musen (Ed), Handbook of child psychology. New York: John
Wiley and Sons.

Maguin, E., & Loeber, R. (1996). Academic performance and delinquency. Crime and
Justice: A Review of Research, 20, 145-164.

Mahoney, J. L., Cairns, R. B. (1997). Do extracurricular activities protect against early
school dropout? Developmental Psychology, 33 , 241-253.

McCombs, A., & Forehand, R. (1989). Adolescent school performance following

parental divorce: Are there family factors that can enhance success? Adolescence,
24, 871-880.

McDill, E. L., Natriello, G., & Pallas, A. M. (1986). A population at risk: Potential
consequences of tougher school standards for student dropouts. American Journal
of Education, 94, 135-181.

McNeal, R. B. (1997). High school dropouts: A closer examination of school effects.
Social Science Quarterly, 78, 209-222.

Morris, 8., Pawlovich, W., & McCall, D. (1997). Evaluating the effectiveness of school
dropout prevention strategies: Some suggestions for future research. Ottawa:
Canadian Education Association.

Murrell, S. A., & Norris, F. H. (1983). Resources, life events, and changes in

psychological states: A prospective framework. American Journal of Community
Psychology, 11, 473-491.

O’Keefe, D. J., & Jensen, J. D. (2005). Do loss-framed messages engender greater
message processing than do gain-framed messages? A meta-analytic review.
Paper presented at the National Communication Association Convention in
Boston, MA.

72

Okun, M. A., Sandler, I. N., & Baumann, D. J. (1988). Buffer and booster effects as
event-support transactions. American Journal of Community Psychology, 16 (3),
435-449.

Persuad, D., & Madak, P. R. (1992). Graduates and dropouts: Comparing perceptions of
self, family, and school supports. Alberta Journal of Educational Research, 38,
235-250.

Plug, E. (2002). How do parents raise the educational attainment of future generations?
NWO Priority Program on Schooling, Labour Market Performance and Economic
Develpment. Department of Economics, University of Amsterdam.

Roderick, M. (1993). The path to dropping out: Evidence for intervention. London:
Auburn House.

Rollins, B. C., and Thomas, D. L. (1979). Parental support, power and control techniques
in the socialization of children (pgs. 317-364). In W. R. Burr, R. Hill, I. Nye, and
I. Reiss (Eds) Contemporary theories about the family (vol. 1). New York: Free
Press.

Rosenfeld, L. B., Richman, J. M., & Bowen, G. L. (2000). Social support networks and
school outcomes: The centrality of the teacher. Child and Adolescent Social Work
Journal, 1 7 (3), 205-226.

Rosenthal, B. S. (1998). Non-school correlates of dr0pout: An integrative review of the
literature. Children and Youth Services Review, 20, 413-433.

Rumberger, R. W. (1983). Dropping out of high school: The influence of race, sex, and
family background. American Educational Research Journal, 20, 199-220.

Rumberger, R. W. (1987). High school dropouts: A review of issues and evidence.
Review of Educational Research, 5 7, 101 -121.

Rumberger, R. W. (1995). Dropping out of middle school: A multilevel analysis of
students and schools. American Educational Research Journal, 32, 583-625.

Rumberger, R. W., Ghatak, R., Poulos, G., Ritter, P., & Dornbush, S. M. (1990). Family
structure on dropout behavior in one California high school. Sociology of
Education, 63, 283-299.

Samter, W., Burleson, B. R., & Murphy, L. B. (1987). Comforting conversations: Effects
of strategy type on evaluations of messages and message producers. Southern
Speech Communication Journal, 52, 263-284.

Sandler, I. N. (1981). On bufl'ers and boosters: Social support for negative and positive

life events. Paper presented at the annual meeting of the American Psychological
Association, Los Angeles.

73

Sarason, I. G., Sarason, B. R., Keefe, D. B, Hayes, B. E., & Shearin, E. N. (1986).
Cognitive interference: Situational determinants and traitlike characteristics.
Journal of Personality and Social Psychology, 51, 215-226.

Schwarzer, R., & Leppin, A. (1991). Social support and health: A theoretical and
empirical overview. Journal of Social and Personal Relationships, 8, 99-127.

Sinclair, M. F ., Christenson, S. L., Evelo, D., & Hurley, C. (1998). Dropout prevention
for high-risk youth with disabilities: Efficacy of a sustained school engagement
procedure. Exceptional Children, 65, 7-21.

Singleton, R. A., & Straits, B. C. (1999). Approaches to social research, 3rd edition (pg.
163). New York: Oxford press.

Smith, S. W., & Ellis, J. B. (2001). Memorable messages as guides to self-assessment of
behavior: An initial investigation. Communication Monographs, 68, 154-168.

Sprecher, S., Metts, S., Burleson, B., Hatfield, E., & Thompson, A. (1995). Domains of
expressive interaction in intimate relationships: Associations with satisfaction and
corrrrnitrnent. Family Relations, 44, 203-210.

Stone, A. A., & Neale, J. M. (1984). Effects of severe daily events on mood. Journal of
Personality and Social Psychology, 46, 137-144.

Strom, R. E., & Boster, F. J. (2005). Dropping out of school: A meta-analysis assessing
the effect of messages in the home and in school. Unpublished paper.

Sullivan, C. F. (1996). Recipients’ perceptions of support attempts across various
stressful life events. Communication Research Reports, 13, 183-190.

Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of
choice, Science, 211, 453-458.

Vallerand, R. J ., Fortier, M. S., & Guay, F. (1997). Self-deterrnination and persistence in
a real-life setting: Toward a motivational model of high school dropout. Journal
of Personality and Social Psychology, 72, 1162-1176.

Vinokur, A. D., & van Ryn, M. (1993). Social support and undermining in close
relationship: Their independent effects on mental health of unemployed persons.

Journal of Personality and Social Psychology, 65, 350-359.

Wehlage, G. G., & Rutter, R. A. (1986). Dropping out: How much do schools contribute
to the problem? Teachers College Record, 8 7, 374-392.

74

Whitbeck, L.B. & Gecas, V. (1988). Value attributions and value transmission between
parents and children. Journal of Marriage and the Family, 50, 829-840.

Willis, T.A., & Shriner, O. (2000). Measuring perceived and received social support. In
S. Cohen, L.G. Underwood, & B.H. Gottlieb (Eds.), Social support measurement
and intervention (pgs. 86-135). New York, NY: Oxford University Press.

Wilson, W. J. (1987). The declining significance of race. Chicago: University of Chicago
Press.

Wilson, S. R., & Kunkel, A. W. (2000). Identity implications of influence goals:
Similarities in perceived face threats and facework across sex and close
relationships. Journal of Language and Social Psychology, 19, 195-221.

Wortmann, C. B., & Lehman, D. R. (1985). Reactions to victims of life crisis: Support
attempts that fail. In I.G. Sarason & B. R. Sarason (Eds.), Social support: Theory,
research and applications (pp. 463-489). Dordrecht, The Netherlands: Martinus
Nijhoff.

Wright, J. D., & Wright, S. R. (1976). Social class and parental values for children.
American Sociological Review, 41, 527-537.

75

1“Emitting1111:1111

1
l1
5