_ -‘- —__“‘_.“‘_Mm_—— .‘m _-I..4...__.‘_- THE EFFECTS OF AN INOIVIOUALIZEO SIONEO , BEHAVIORAL CONTRACT BETWEEN A VOLUNTEER WORKER ANO HIS/HER SUPERVISOR: A FIELD _ EXPERIMENT Thesis for the Degree Of M." A. MICHIGAN STATE UNIVERSITY LYNNE CUTLER 1975 9'39. ”’4, :3 ”4‘ ma; ‘ . 4 far”. ‘_....__ I v Lu;13::113-+ty r, .;'\— w—v— —- “5553; \‘Q; Q'Pafl‘ygtu ; NNNNNINNNINN ABSTRACT THE EFFECTS OF AN INDIVIDUALIZED SIGNED BEHAVIORAL CONTRACT BETWEEN A VOLUNTEER WORKER AND HIS/HER SUPERVISOR: A FIELD EXPERIMENT By Lynne Cutler Student volunteers in four social service agencies were randomly assigned to three conditions: 1) a contract conditon where a volunteer wrote an individualized signed behavioral contract with his/her super- visor that specified expected, less than expected and more than expected performance levels; 2) a discussion conditionvhere a volunteer met with his supervisor to discuss general feelings, anxiety and concerns; and 3) a control condition where a volunteer did not have these planned meetings or a behavioral contract. These treatment conditons were tested to see the effect on volunteer's satisfaction, supervisor/teacher ratings, interaction and changes in the child/children the volunteer works with. A multivariate analysis of variance revealed a positive effect due to the contract and discussion conditions on volunteer's, activity and supervisor's satisfaction. The results also indicated that a volunteer's need fbr clarity moderates the effectivenesscf the contract such that a contracted volunteer with a high need for clarity stays with the agency longer, has a greater satisfaction towards his activities and reports a better relationship with his child than a volunteer with a high need for clarity not on the contract. This type of volunteer also tends to be more satisfied with his supervisor and has a clearer role perception. THE EFFECTS OF AN INDIVIDUALIZED SIGNED BEHAVIORAL CONTRACT BETWEEN A VOLUNTEER WORKER AND HIS/HER SUPERVISOR: A FIELD EXPERIMENT BY Lynne Cutler A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS .Department of Social Sciences 1975 TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES . INTRODUCTION . Review of the Literature METHODS Subjects Research Design Measures Procedure RESULTS Background Characteristics Treatment Effects Agency Effects DISCUSSION APPENDICES A. Volunteer Contract B. Volunteer Questionnaire C. Volunteer Evaluation D. Volunteer Progress Report BIBLIOGRAPHY . ii Page iii iv Table DOOM U1 10. ll. 12. l3. I4. 15. LIST OF TABLES Agency Characteristics . Research Design and Sample Size in each Treatment Level . Comparison of Treatment Groups on Background Characteristics Comparison of Volunteers who quit with those who did not, according to Background Characteristics Interaction Effects Treatment Effects: Results of the Multivariate and Univariate Analysis of Variance on Criterion Measures Treatment Effects. Estimates of the Contrasts and the 95 Percent Univariate Confidence Intervals . Intercorrelations Between Activity Satisfaction, Supervisor Satisfaction and Months At Agency Agency Effects: Results of the Multivariate and Univariate Analysis of Variance on Criterion Measures Treatment Effects. Combined Cell Means and Standard Deviations of Criterion Measures . . . Agency Effects: Estimates of the Contrasts and the 95 Percent Univariate Confidence Intervals Agency Effects: Combined Cell Means and Standard Deviations of Criterion Measures Correlations Between Meeting Goals and the Criterion Measures . . . . . . Correlation Between Volunteer Evaluations and Child's Personality Measures Correlations of Need for Clarity and Criterion Measures for Contracted and Non-Contracted Volunteers . iii Page 19 20 22 23 24 25 26 27 29 30 32 33 34 LIST OF FIGURES Figure Page l. Test Administration . . . . . . . . . . . . . . l0 2. Summary of Activities---October l974-May l975 . . . . . l4 iv INTRODUCTION In recent years the utilization of volunteers in mental health agen- cies and schools has increased. Often these organizations place the vol- unteer and agency in a polemic situation where the agency has few guide- lines on how to use the volunteer and consequently, the volunteer is un- clear of his/her role in the agency. Perhaps this confusion contributes to the volunteer's dissatisfaction, dropout rates, and underutilization 'of talents and skills. How then, can an agency provide guidelines to its volunteers? One possible method used to communicate behavioral goals is the goal attain- ment scaling procedure developed by Kiresuk and Sherman (l968). Criteria for expected, less than expected, and more than expected levels of treatment/administrative outcomes are written in a contract format. In- itially, this procedure was designed to evaluate different treatment modes within a mental health agency, but subsequently has been used in settings to evaluate mental health programs; the outcome of treatment and training programs on clients; the use of evaluation feedback data in group therapy sessions, and mutually detennined goals between client and therapist (Garwick and Vanderpool, l973). Similar procedures, such as management-by-objectives (Odiorne, l965), goal-setting (Locke, l966) and behavioral objectives (Duchastel, l973), are used in other formal settings, 1,32, industry, scientific research, and most currently in education. Usually the specific methods used to identify and quantify the goals are not delineated. Mager's goal analySis l procedure is a useful tool in approaching this problem (l972). This procedure consists of five steps whereby goals and expectations that are considered important are translated into specific measurable be- haviors. Ideally, it allows one to discover the essence of a goal so it can be communicated and evaluated objectively (Mager, l962, 1972). The agencies selected for the study have general objectives and goals, an accomplishment few volunteer programs have. Most of these goals and objectives, however, are not tangibly related to the volun- teer's skills and behaviors and therefore may become statement of in- tent rather than the realization of those objectives. Perhaps then the tecnniques of goal analysis and goal attainment can be used with volunteers to decrease dropout rates and increase satisfaction and ef- fectiveness. Specifically, then, this study assesses the effectiveness of an individualized, signed behavioral contract between a volunteer worker and his/her supervisor. Review of Literature Goals guide a person‘s actions and thoughts. Locke maintains that "peOpleis statements about how they intend to perform and their goals in a situation are good predictors of their actual performance in the situation." (Cited by Lawler, l970.) This suggests that when people set goals for themselves, they are motivated to act according to these goals. For example, Bryan and Locke (l967), in a laboratory setting, found that when an initially slow-motivated group were given Specific goals they caught up with an initially high-motivated group. It was also found that setting goals in boring tasks increased interest (Locke, l967). In a later study, Locke (l968) found persons with very hard goals perform consistently better than persons with easy goals, even though the former 3 may not realize their goals; and in his review of eight laboratory studies, six snowed that persons with highly specific goals performed at significantly higher levels than persons with the more general goal of "doing their best." Further, Locke and Bryan (l969) found that a person's goals affect not only how hard a person works, but also what aSpects of a more complicated task he focuses on. Recently, Latham and Baldes (l975) empirically tested LockePs theory in six lOgging companies using a time series design. They found that performance improved im- mediately upon the assignment of a specific hard goal as opposed to the more generalized goal of "doing your best," and that this increase held constant across nine consecutive months, the duration of the study. Management-by-objectives is a popular assessment technique used in industry that combines feedback performance with goal-setting. In l965 Meyer, Kay and French empirically studied the effects of a participa- tive interview with mutual goalsetting versus a non-participative inter- view with superior-imposed goals. Results showed that mutual goal-setting resulted in more goal accomplishment tnan super-imposed goals; and per- sons accustomed to participation reacted more favorably to the participa- tive interview than those not accustomed to participation. A person, however, may be very motivated and set goals for a parti- cular job or task and still perform incorrectly or become disenchanted because he perceives his role incorrectly or is subjected to incongruent messages. Role conflict exists when the behaviors expected of an indi- vidual are inconsistent and the person experiences stress, becomes dis- satisfied and performs less effectively than if the expectations imposed on him did not conflict (Kahn, §§¢_gl,, l964). For example, conflict exists when a person's internal standards and values are in Opposition 4 to the defined role behaviors. A theoretically and empirically sep- arate construct-role ambiguity refers to the lack of necessary infor- mation available to a given organization position that increases the probability a person will be dissatisfied with his role, will exper- ience anxiety, will distort reality, and will thus perform less effec- tively (Kahn, l964). Rizzo gt;_gl, (l970) reviewed several articles that provide empirical support to role ambiguity theory, and conclude that role ambiguity results in undesirable consequences for both or- ganization members and for organization performances. Similarly, Law- ler (l970) maintains that it is important for supervisors and subor- dinates to have common perSpectives so that the subordinate does not perform incorrectly because he perceives his role inaccurately. He suggests that supervisors and subordinates work toward developing shared perceptions, i;g., Specific goals and objectives of how the subordinate's job should be accomplished. To alleviate ambiguity of the volunteer's role, John Cauley, past director of the Office of Volunteer Programs at Michigan State University, suggests that agencies write volunteer job descriptions that outline the volunteer's duties and reponsibil- ities (l974). Further, Sorum (l974) maintains that behavioral goals can and should be used at various levels of a volunteer program. For instance, they can be used to Specify agency, individual staff members', or individual volunteers' roles and objectives. In general, there is evidence that "goal-setting" is a beneficial tool that provides the agency with evaluative feedback and may increase performance. It is possible, however, that individual differences may diminish or enhance the contract's effectiveness. For example, in a survey of the literature, there are some personal variables that seem 5 to predict the volunteer's effectiveness. Goodman's study on big brothers (l972) reports that a college student with an outgoing profile was the most effective companion therapist. Even though this profile is mainly a composite of various personality tests, including a ninety- minute group assessment procedure, specific background characteristics also emerged. The person who had previous experience with children, a person-oriented college major, had participated in many extracurricular activities during high school and college, with a high school grade point average around 3.0 was most effective. Mitchell (l966) found that students who dressed casually, were idealistic, spontaneous, enthusiastic, and dependable were most effec« tive, while those in their first term at school tended to overestimate the amount of free time they would have available and thus, were poor risks. Further, Truax and Mitchell (l97l) in a review of salient ther- apist characteristics concluded that empathy, warmth, and genuineness are most important. In their paper, they also outline a procedure to select helping personnel with these characteristics. Specifically, they look for people low in anxiety, depression, introversion who at the same time are striving, dominant, active autonomous individuals. A variable specifically related to the present research design con- cerns a person's need for clarity; i;§,, how does this need interact with his job satisfaction and effectiveness? In an experimental study, Lyons (l97l) investigated the relationship between role clarity and re- ported satisfaction, tensions, prOpensities to leave and withdrawal of registered nurses having different needs for clarity. The results showed that: 1) when nurses demonstrated a high need for clarity, role ambi- guity significantly decreased satisfaction, increased tensions, and 6 increased propensities to leave; 2) when nurses showed a low need for clarity, there were no significant correlations with these four factors; and 3) the low need for clarity nurses were not upset when they saw their roles as structured, which suggests that one can increase the degree of role clarity without driving out or dissatisfying those with a low need for clarity. In general, the volunteer literature does not address the issue of goals as it pertains to volunteer organizations. This in turn will not alleviate the aforementioned ambiguous situation between the volunteer and the agency. In light of these issues, an experimental application of behavioral goals in volunteer agencies should be helpful to deter- mine the usefulness of setting a behavioral contract. To this end, this study examines the efficacy of a behavioral contract between a student volunteer and a volunteer coordinator for social service agen- cies working with children and adolescents. In contrast to other studies where treatment effects are aimed primarily at the volunteer's development (Gruver, l97l), this study also measures the effects of the treatment on the children. Specifically, then, the following hy- potheses tested tne effects of behavioral contracting on volunteers and the children they worked with: l) The contracted volunteer will have greater job satis- faction than non-contracted volunteers. 2) The supervisors and teachers will evaluate contracted volunteers as more effective. 3) The contracted volunteers will be more effective with the children. 4) The contracted volunteers who attain their goals will have greater job satisfaction, higher supervisor ratings, and will be more effective with the children than the volun- teers who did not attain their goals.. 5) The volunteers with high need for clarity will be more effective with the contract than without. In addition, correlational data on the relationship between volun- teer characteristics (demographic and personality) and the volunteer and child outcome measures are also collected. However, no fonnal hypotheses are presented. METHODS Subjects Seventy-seven (34 male and 43 female) single Michigan State Uni- versity students participated in this study. They ranged in age from l8 to 29 with a mean age of 20. Of these, 74 were white. These stu- dents volunteered to MSU Office of Volunteer Bureau and were placed by November l, l974, into one of the following agencies: Big Brother/Big Sister of America, Everett Big Brothers, Reach-~Everett Elementary School, and Gier Park Elementary School. These agencies are located in Lansing, Michigan, and require the volunteer to meet individually with a child once a week for nine months. Most of the children find it difficult to adjust socially, academically or emotionally to their environment. Table l shows the initial sample size, target pOpula- tion, and the type of conditions under which the volunteer worked. By March T975, 15 volunteers quit and two children moved. Table l. Agency Characteristics Agency # of Volunteers Target Population Working Environ. Big Brother/Big Sister 33 6-l7 years old varies Everett Big Brothers l3 elementary after schl. Reach l9 elementary in & after school Gier l2 elementary-- in school learning disability Research Design This is a 4 X 3 factorial design, with 4 agencies and 3 treat- ment conditions. Some measuring instruments were administered in a pretest-posttest format, although for some only a posttest was ad- ministered. Table 2 illustrates the design and the sample sizes. Within each agency half the volunteers were assigned to the behavioral contract condition. In this condition the supervisor wrote an individualized signed contract with the volunteer (Appendix A) that specified expected, less than expected, and more than expected performance levels. The vol- unteer also wrote personal goals and expectations. The remaining volun- teers were divided into two groups in order to determine if the contract worked because of the contract specifically or as a result of increased rapport between the Volunteer and the supervisor: a discussion condi- tion where the volunteer met with the supervisor to discuss general feelings, anxieties, concerns; and secondly, a control group where the volunteer did not have these planned meetings or a behavioral contract. Because of Gier Park's small sample size, it did not have the discussion condition. Table 2. Research Design and Sample Size in each Treatment Level Agency Contract Discussion Control (DI~_ (n) (n) Big Brother l7 8 8 Everett Big Brothers 5 4 Reach 9 ’ 5 5 Gier Park . 6 O 6 10 Big Brother/Big Sister volunteers were randomly assigned to the treatment conditions. In the other agencies, the volunteers attended on the same day and thus, days were randomly assigned and counterbal- anced across agencies in order to keep the conditions separate. Although there is no reason to suspect any systematic difference in the days the students volunteered, demographic information and personality measures were initially taken on all volunteers. And as can be seen in the re- sults sections, the treatment groups did not differ on any of these mea: sures, which supports the randomization's effectiveness. Measures In selecting criteria for change, the experimenter attempted to in- clude measures that were standardized, brief, and sensitive to slight changes. All had good internal consistency (r > .60) in the current study using Hoyt Analysis of Variance. The following describes the covariate indices (volunteer background measures), and the criterion measures (volunteer satisfaction, volunteer evaluation, process mea- sures, and children measures). Figure 1 shows when the tests were ad- ministered. Figure l. Test Administration October December March April/May Background Questionnaire X Volunteer Satisfaction X Volunteer Evaluation X I X Process Measures ' X Children Measures . X X ll Background Questionnaire and Personality Measures (see Appendix B) Each volunteer was asked questions concerning his/her religious back- ground, religiosity, ethnic background, parental occupation and educa- tion, college major, reasons for volunteering and previous work exper- iences with children. Another set of questions tapped the person's need for clarity (Lyon l97l), trust in people (Survey Research Center 1966), empathy towards others (Hogan 1969), and personality in terms of extra- version-introversion and neuroticism-stability (Eysenck Personality In- ventory Format l968). The need for clarity scale consisted of four questions each with five alternatives. Scores can range from 5 to 20. Split-half relia- bility was estimated to be .82; in the current study, r is .79. Trust in people (Survey Research Center l969) is a three forced-choice scale that measures a person's trust in others. Scores range from T (low trust) to 6 (high trust). Inter-item correlations were at least .48. For this study, internal reliabilities were .72 and .61 for the volun- teers and children respectively. The empathy scale (Hogan 1969) is a 64-item, forced-choice pencil and paper test that measures the ability of the person to understand another's condition or state of mind without actually experiencing that person's feelings. In the scale's construc- tion, internal consistency coefficient using the Kuder-Richardson for- mula Zl is .7l and for test-retest correlation after two months, r= .84. The split-half reliability for the Eysenck inventory ranged from .74 to .9l using Spearman Brown prophecy formula and test-retest reliabilities range from .80 to .97 for the separate test forms. 3 Volunteer SatisfactiongQuestionnaire. Two scales from the Job Description Inventory (Smith, Kendal, Hulin, l969) were used as depen- dent measures. One scale measured the satisfaction with fjob" activities; l2 the other measured the volunteer's satisfaction with his supervisor and/or teacher. Beside each item the volunteer was asked to write "Y" if the item described a particUlar aspect of his activities (supervi- sion), "N" if the item did not describe that aspect,.or a "?" if he could not decide. Scores can range from 0 to 54. "V" was scored 3, "N" was scored 0, and "7" was l. The estimated split-half internal consistency corrected by the Spearman Brown formula was .84 for activities and .87 for the supervision scale. For the volunteer's satisfaction with his activities, r = .79; for volunteer's satisfaction toward his teacher, r = .90. The role ambiguity scale designed by House (l970) was used to mea- sure the volunteer's understanding of his duties. The volunteer rated himself on six items, using a seven-point scale ranging from very false to very true. House reports internal consistency estimates of .78 and .80 for these six items. Internal reliability for the volunteers in this study is r = .84. A record was kept on the number of months each volun- teer stayed in the agency. Volunteer Evaluation.. In review of the literature, no standard- ized evaluation forms appropriate to the activities of the volunteers in this study were found to measure the volunteer's performance. As'a re- sult, the experimenter designed a thirteen-item, five-point scale (Appen- ciix C) that assessed volunteer behaviors that seemed appropriate to the Irolunteer's responsibility. Scores can range from 0 to 52, high scores r~eflecting higher performance. This scale has good internal consistency estimates for the supervisors, r = .92 and .96, as well as r = .93 and .90 for the teachers. 13 Process Measures. The volunteer described his overall relation- ship with the child, using Goodman's Report for parts 5 and 6 (l972). Part 5 describes the amount of intimacy witnessed by the volunteer during the time spent together, and part 6 contains ten items on a six-point scale relevant to a two-person encounter. In this section, first the volunteer described his feelings or behaviors and then de- scribed his perception of the child's feelings or behaviors during their visits. No internal consistency reliabilities were reported for the scale; however, in this study r is .94 for the volunteer's interaction and .92 for the volunteer's report of the child's interaction. Children Tests. Since ultimately the volunteer's effectiveness is reflected in the children they work with, five change measures were at- tained on the children. 1. The teacher or school counselor rated each child's self- concept using McDaniel's Inferred self-concept scale (l969). This thirty-item, five-point scale mainly covers social relations. Scores can range from l to 5, low scores reflecting higher performances. Re- ported split-half reliabilities were .86, .86, and .90 for ratings by counselors, teachers, and both combined. The internal consistency es- timate for the participants in this program is .92. 2. The child's trust in people (Survey Research Center, l969). 3. Grades, absenteeism and tardiness record. Procedure Figure 2 outlines the steps taken in the study. 4 .upwgo we pcmscm_mmm\Aucmmm um cowmmmm :owumpcmmco .cmcv_wgu Co mcwpmmpmcm .mgwmccowp Immzc uczccmxumn meuczpo> .mzasmo co cowumpcmvco azocw .mmwucmmm gpwz magma -mmcmm pmspomcucou umcwmpno cowuwucou Foepcou .A.::_o> Lma .cws m_umv cmmpca I_o> on“ Co mccmucou Lo .mmwum wacm .mm:__mm¥ xcm Fmemcmm :_ mmzumwu 0» new no: m>mg Xme cow Icon mca mcowummzc xcm cmzmcm cg mgmmp::_o> zumpumcpcouucocz umpmcmwmmu guwz qu Lomw>cmazm .n_w;u mo “smacmwmmm\xocmmm pm cowmmmm cowpmucmwgo .cmcu__cu Co mcwpmmumca .mcwmccowu Immzc uczogmxomn emwp::_o> .m3QEmu co :owumpcmwgo gzoeo .mm_ucmmm saw; mpcms Immcmm szpumcpcou nonwmuno Apumeem onmumfiav :o_pwucou cowmmzumwo Agmmpczpo> emu .:PE mp Imv .uumcpcou PmLow>mcma on» muwcz op mcmmp::_o> =umpomcp Taco: :u_z awe Lomw>cmazm Acme; mI—v .mmmuoga mwcb cw comw>cma=m :_meu on use .xu:mmm L_m;u cw mFOL m.cmmp::Fo> mg» co mwmAFm Icm Pmom m Ecomcma op Lom_> Igmasm :pwz me cmpcmswgqum .u_w;u we ”cmscmwmmm \xocmmm “a cowmmmm cowumucmwco .cmcu__:u Co mc_ummpmcm .mc_m::owp Immsc uczocmxumn coupes—o> .mqumo co :o_bmpcmwco azoeo .mmwucmmm :pwz mpcme Immcmm Fmauumeucou umcwmuao cowpwucoo Fmpcmswcmmmm mxmp am2-4kmp caseboo---mmwpw>_bu< Co NMSEE=W .N mczm_m LmDOpuo LmnEmuamm mgpcoz * mmbm 15 .mewweeowm Immec eowpmwemwpwm emmpe:_o> .emewe_eu eo “wwwwmoe A.E oFImv emmme:_o> mew mew: IPw>m ememwmu new eomw>emeem A.E o_Imv emmmenpo> mew mew: -_w>m ememwmp new eomw>emesm cowbeweoe Foeeeom .mewweeo_p Immec eowpmwemwpwm emmue:_o> .emen_eeu eo pmmewmoe A.E QFImv emmpeepo> mew mew: upw>m ememwmu new enmw>emesm .A.enpo> eme .E mFImV emmpe: IFo> men we meemmeom eo .mmwp Imwxew .mmewpmme zew mmemm_n on new nwe m>we awe enmeme mew meowpmmec eew em2mew op memmu Ie:_o> =nmpmwepeomueoe= nmmwe Immmmn epwz pms eomw>emeem A.E OFImV emmpe:_o> mew mew: IFw>m ememwmp new eomw>emeem A.eepo> eme .E mFImV .emmue: I_o> mew mo meemmeou eo .mm_p Im_xew .mmew_mme mew mmemmen on new nwe m>we wa eomeme men meoemmmnc mew emzmew on memmp Tee~o> =nmpuweueomueoe= nmuwe Imwmmn ewe: umE eomv>emenm Nuumeem ammuweev eowuwneou eowmmnmmwo .mewweeowp Immec eowpuwemwpwm emmpe:_o> .emeweeeu e0 wmwwwmoe A.E OFImV emmpenFo> mew mew: I—w>m emeowmp new eomw>emenm .Aemmpee—o> eme .ewE mpumv umweeeom men mmw>me on new meom pmwepeom meu mnewzou mmmem Ioea m_emmpe:_o> me» 3mv> -me on memmue:_o> :nmpmwem neon: ewe; ems eomw>emezm A.E o_umv emmuez—o> men mmw: I_w>m ememwmp new eomw>emaem .Aemmpe:_o> eme .ewe mFImV umwepeom mep mmw>me on new m_wom umweueom mew mnewzom mmmem -oee m.emmme:_o> mew 3mw> Ime op memmpenpo> =nmpmweu Ieomz new: uma eomw>emezm eowpwneou premewemQMm we ewz\_eee< we NP euewz pp op emeEmmmo m mmuwo * empm .A.n.ueouv N menmwe 16 Typically, volunteers at Michigan State University contact the MSU office of volunteer programs and are referred to an agency that inter- ests the student. Then, the agency holds an orientation meeting and some screen the volunteers. (0f the four agencies selected for the study only Big Brother screens its volunteers.) After this, the volunteers are assigned to their child, teacher, and other responsibilities. In this study, all the volunteers followed the above procedure, except that they were also asked to complete a background questionnaire.. The volun- teers were told that a graduate student in conjunction with the Office of Volunteers Programs were evaluating some programs and trying to deter— mine whether certain volunteers worked better with certain children. The volunteer did not know of the contracting experiment. Then, as previously explained, the volunteers were assigned to the treatment conditions. First, the supervisor met with the discussion group. This was done in order to keep the contract and discussion (pla- cebo) conditions separate because the supervisor met with both the exper- imental and placebo groups. Next, the experimenter met individually with the supervisors to perform a goal analysis on the volunteer's role in their agency, and to train the supervisors in this process. Then in December and March, the supervisor met with the contract and discus- sion groups (Big Brother/Big Sister did not meet with the volunteers in December.). This time, however, when the supervisor met with the contracted volunteers they jointly evaluated the volunteers' progress towards the previously specified goals and, when necessary, revised the contract. (See Appendix D for Volunteer Progress Report form.) Thus, volunteers in the contractual group and discussion met the same number of times. (Three times for all agencies except Big Brothers, which met 17 twice.) Control volunteers had no scheduled contact with their super- visors. In all other aspects of the volunteer program, the volunteers were treated the same. Actual contact between vOlunteer and child ranged from 30-45 min./wk. at Gier Park, from l-3hrS./wk. at Reach, from T l/2-3 hrs./wk. at Everett, and 2—6 hrs./wk. at Big Brother/Big Sister. RESULTS Background Characteristics To address the issues concerning the efficaCy of the randomization and any biases that could be attributed to attrition, differences in treatment conditions versus quit-did not quit were tested in two-way ANOVA's and MANOVA'S on the volunteer's background characteristics. The interactions were not significant on any of the background variables (p < .24). And as can be seen in Table 3, the multivariate F ratios indicate the treatment groups are homogeneous with respect to age, number of children in the family, birth order, SES (Occupational Code/Wisconsin Bureau of Census, T963), population of hometown, college year, number of activities in college and high school, reasons for volunteering, per- sonality factors, religiosity. Also, the volunteers did not differ in sex (x2 = .59, df = 2, p < .198); college major (x2 = 16.07, df = 10, p < .10); previous experience with youth (x2 = .54, df = 2, p < .76); elected offices (x2 = 4.39, df = 2, p < .11); high school GPA (F(2,65) = 1.67, p < .19); or college GPA (F(2,58) = 1.66, p < .20). Similarly, those who quit were no more likely than those who didn't quit to differ in sex (x2 = 1.54, df = 2, p < .46); college major (x2 = 3.58, df ; 4, p < .46); previous experience with youth (x2 = 2.15, df 2, p < .34); elected office (x2 = .59, df = 2, p < .74); high school GPA (F(l,65) = 1.53, p < .22); or college GPA.(F(l,58) = 3.67, p < .06). Furthermore, the hypothesis for no main effects due to quitting could not be rejected on any of the other variables asillustrated in Table 4. 18 19 Table 3. Comparison of Treatment Groups on Background Characteristics Variable N Contract Discussion Control 'Univariate Multivariate X X X F p< F ~p< Age in yrs. 20.1 19.9 20.05 .09 .91 # of siblings 3.67 3.12 3.7 .49 .61 Birth order 2.1 4.2 2.2 1.29 .28 SES 58. 58.1 51.2 .46 .63 Hometown pop. 71 3.3 2.9 3.3 .78 .46 .52 .88 Yr. in coll. 2.81 2.47 2.87 .58 .56 # coll. activ. 2.54 5.06 2.13 2.19 .12 # h.s. activ. 3.81 7.3 4.04 3.78 .03 # reasons for volunteering 3.2 2.9 2.95 .05 .95 # academic reasons 77 .95 76 .83 .09 .91 1.06 .39 Neuroticism 8.2 . 8.5 8.3 .07 .94 Extroversion 10.9 13.9 14.4 1.86 .16 Trust 5.28 5.53 5.55 .58 .56 Empathy 40. 41.5 39.1 .86 .43 Need for clarity 75 14.7 12.5 14.45 3.03 .055 1.48 .15 Religiosity 2.8 2.6 2.9 .29 .75 Church atten- dance 72 2.7 2.65 2.4 .6 .55 '1.09 .36 20 Table 4. Comparison of Volunteers who quit with those who did not, according to Background Characteristics Variable N Did Not Quit Quit Univariate Multivariate X X F p< F p< Age in yrs. 20.09 19.79 .26 .61 # of siblings 3.46 3.93 .69 .41 Birth order 2.67 2.50 .01 .90 SES 55.37 58.57 .15 .69 Hometown popul. 71 3.26 3.07 .32 .58 .38 .86 Yr. in coll. 2.81 2.53 .51 .48 # coll. activ. 3.07 2.60 .12 .73 # h.s. activ. 4.37 5.8 1.1 .30 # reasons for volunteering 2.89 3.73 2.33 .13 # academic reas. 77 .79 1.2 1.55 .22 1.83 .12 Neuroticism 8.1 9.0 .57 .45 Extroversion 12.72 12.07 .09 .76 Trust 5.47 5.2 .94 .34 Empathy 40.20 39.60 .14 .71 Need for clarity 75 14.12 14.27 .03 .87 .34 .89 Religiosity 2.74 2.8 .04 .84 Church atten- dance 72 2.57 2.6 .004 .95 .02 .97 To summarize, these analyses show no systematic drop-out rate by treatment group and serve as a justification of the treatment group's homogeneity. Before preceeding to the results of the dependent measures, some comment is in order concerning the manner in which the data were analyzed. The first three hypotheses were tested with multivariate and univariate analyses of variance. In general, when the criteria seemed to measure the same construct, a multivariate analySis, agency by treatment, was em- ployed. However, sample sizes on some of the measures differed because a 21 volunteer quit, or there were no supervisor-teacher evaluations for a par- ticular agency. As a result, the appropriate univariate analysis, agency by treatment, was used for the evaluation measures, and for determination of the volunteer's satisfaction with his/her teacher. Since this is a nonorthogonal design, unequal cell size, and the aim was to test each main effect, a reordering was done for each criterion measure when either of the main effects was significant. Therefore, the probabilities given in the text refer to the main effect as it was or- dered last in the MANOVA or ANOVA run. Lastly, in the multivariate ana- lyses, the univariates will not be discussed unless the multivariate F ratio is significant at the .10 level. The analyses for the first, second and third hypotheses that con- tracted volunteers will be more satisfied, have higher evaluations and be more effective, are shown in Tables 5, 6 and 9 and the treatment and agency observed combined cell means are reported in Tables'HJand 12 respectively. None of the interactions between agency and treatment are significant (Table 5). Treatment Effects Table 6 presents the results of the main effect due to the treat- ment conditions: contract, discussion, control. 22 Table 5. Interaction Effects Univariate' Multivariate ' Variable N df F p< F P< VOLUNTEER MEASURES Activity Satis. 66 5,55 1.18 .332 Supervis. Satis. 66 5,55 .82 .54 Months 66 5,55 .38 .86 Multivariate Satis. 66 15,146 .72 .76 Teacher Satis. 24 1.19 1.79 .196 Role Clarity 66 5,55 .82 .543 Teacher Eval. Fall 27 1,23 .41! .53 Teacher Eval. Winter 19 1.14 3.17 .097 Super. Eval. Winter 62 5,51 .34 .889 PROCESS MEASURES Vol. Interaction 67 5,56 1.01 .42 Child Interaction 67 5,56 2.23 .064 Multivariate Inter. 67 10,110 1.27 .25 CHILD MEASURES GPA Spring 52 5,44 1.13 36 Absent 52 5,44 1.44 .23 Tardy 52 5,44 .89 .49 Multivariate Acad. 52 15,116 1.12 .35 Self-Concept 67 5,56 .86 .52 Trust .49 5,38 .31 .907 23 Table 6. TREATMENT EFFECTS: Results of the Multivariate and Univariate Analysis of Variance on Criterion Measures ' Univariate Multivariate Variable N df F p< F p< VOLUNTEER MEASURES Activity Satis. 66 2,55 3.62 .033 * Supervis. Satis. 66 2,55 3.66 .032 * Months 66 2,55 .92 .40 Multivariate Satis. 66 6,106 2.0 .071 + Teacher Satis. 24 2,19 .82 .456 Role Clarity 66 2,55 .81 .452 Teacher Eval. Fall 27- 1,23 .00 .977 Teacher Eval. Winter 19 2,14 1.82 .198 Super. Eval. Winter 62 2,51 2.19 .122 PROCESS MEASURES Vol. Interaction 67 2,56 1.62 .208 Child Interaction 67 2,56 3.44 .039 - Multivariate Inter. 67 4,110 1.69 .157 CHILD MEASURES GPA Spring 52 2,44 .87 .43 Absent 52 .2,44 2.33 .11 Tardy 52 2,44 1.18 .32 Multivariate Acad. 52 6,84 1.39 .229 Self-Concept 67 2,56 .45 .64 Trust 49 2,37 .38 .68 +p<.IO * p < .05 24 The only significant treatment effect occurred in the volunteer satis- faction measures (Multivariate F = 2.01, df = 6.106, p < .0713). To further explain the significant treatment effects, the univariate F statistics were examined to locate which of the individual variable contributed most to the observed multivariate effects and are also found in Table 6. An inspection of the univariate F statistics indicates the volunteer's satisfaction and activity satisfaction are the primary contributors (p < .035). In order to determine the nature of the significant treatment ef- fects, the estimated differences between the following mean contrasts were examined: 1) contract minus discussion; 2) contract minus other; 3) contract plus discussion minus other; 4) contract minus control. These estimates and their 95 percent univariate confidence intervals are given in Table 7, following Yelon and Schmidt's (1973) procedure. Table 7. TREATMENT EFFECTS: Estimates of the Contrasts and the 95 Percent Univariate Confidence Intervals Contrast Activ. Satis. Super:Satis. Contract minus Discussion -1.46 i 5.36 .09 i 5.28 Contract minus Other 2.36 i 4.23 3.39 e 4.16 Contract plus Discussion minus Control 6.92 H- 5.05 * 6.66 4.97 * H- + Contract minus Control 6.19 _ 5.14 * 6.71 113.14 * Examination of these contrasts shows the first and second are not significant. Thus, the predicted hypothesis that the contracted 25 volunteers would be more satisfied is not supported. The third and fOurth contrasts, however, are significant and positive, suggesting that persons who had more contact with their supervisors are more satisfied with their supervisors and activities. The intercorrelation matrix between the three dependent measures-~- activity satisfaction, supervisor satisfaction, and number of months in the program---reveals an interesting pattern, as shown below in Table 8. Table 8. Intercorrelations Between Activity Satisfaction, Supervisor Satisfaction and Months At Agency A S M Activities (A) 1.00 (Symmetrical) Supervisor (S) .58 ** 1.00 Months (M) .34 ** .18 1.00 ** p <.O1 A correlation of .58 was found between the activity and supervisor measures. However, supervisor satisfaction is not correlated with months, whereas activity satisfaction is significantly related with months. What this suggests is that reported activity satisfaction is a better predictor of actual behavioral outcomes. AgencygEffects The multivariate F ratio for the main effect due to agency is also significant for the satisfaction measures. 26 Table 9. AGENCY EFFECTS: Results of the Multivariate and Univariate Analysis of Variance on Criterion Measures Univariate Multivariate Variable N df F p< F p< VOLUNTEER MEASURE Activity Satis. 66 3,55 .94 .427 Supervis. Satis. 66 3,55 .52 .668 Months 66 3,55 5.33 .003 ** Multivariate Satis. 66 15,146 2.14 :025 * Teacher Satis. 24 1,19 .04 .85 Role Clarity 66 3,55 1.17 .331 Teacher Eval. Fall 27 1,23 3.51 .074 Teacher Eval. Winter 19 1 14 .87 .368 Super. Eval. Winter 62 3:51 8.01 .0002 ** PROCESS MEASURES Vol. Interaction 67 3,56 1.64 .189 Child Interaction 67 3,56 .72 .544 ' Multivariate Inter. 67 6,110 1.45 .207 CHILD MEASURES . GPA Spring 52 3,44 .69 .565 Absent 52 3,44 .85 .476 Tardy 52 3,44 .87 .465 Multivariate Acad. 52 9,102 .77 .65 Self-Concept 67 3,56 .72 .55 Trust 49 3,37 1.16 .34 p < .05 'k * p < .01 Table'ML TREATMENT EFFECTS: Combined Cell Means and Standard Deviations of Criterion Measures Variable Contract Discussion Control VOLUNTEER MEASURES Activity Satis. X 43.97 44.93 37.94 SD (6.00) (5.11) (11.93) N 33 15 18 Supervis. Satis. X 47.85 47.67 40.67 SD (6.73) (4.59) (10.34) N 33 15 18 Months X 6.05 6.7 6.45 . SD (1.54) (1.18) (1.45) N 37 17 23 Teacher Satis. X 46.27 38.25 43.78 SD (6.34) (5.19) (14.41) N 11 4 9 Role Clarity X 25.15 24.0 22.89 SD (4.98) (2.94) (7.82) N 33 15 18 Teacher Eval. Fall X 39.17 39.8 SD (12.19) (10.64) N 12 15 Teacher Eval. Winter X 38.75 45.5 45. SD (7.83) (6.24) (7.29) N 8 4 7 Super. Eval. Winter X 34.97 41. 31.18 SD (10.58) (7.39) (6.71) N 30 15 17 PROCESS MEASURES - Vol. Interaction 48.85 51.36 49.65 X (4.79) (3.81) (4.27) SD 33 14 20 N Child Interaction . X 45.3 50.00 47.2 SD (6.48) (4.26) (4.93) N 33 14 20 28 Table 10.--Continued Variable Contract Discussion 'Control CHILD MEASURES GPA Spring X 2.33 2.61 2.06 SD (.82) (.875) (.83) N 27 13 16 Absent X 2.79 7.14 2.47 SD (1.01) (8.53) (.95) N 27 13 16 Tardy X 38.15 28.08 46.88 SD (33.81) (25.3) (20.68) N 27 13 16 Self-Concept X 4.02 4.15 3.67 SD (2.31) (1.911) (2.24) N 32 15 20 Trust X 4.96 5.27 4.92 SD (1.04) (.521) (.86) N 25 11 13 29 Examination of the univariates, however, reveals that this effect is only significant with respect to months (p < .0028). Three agency comparison groups were formed: 1) Big Brother/Big Sister minus Everett; 2) Reach minus Gier; 3) Big Brother plus Everett minus Gier plus Reach, as can be seen in Table 11. Table 11. AGENCY EFFECTS: Estimates of the Contrasts and the 95 Percent Univariate Confidence Intervals Contrasts . Supervisor Evaluation Winter Months BB/BS minus Everett -9.74 i 7.33 * 1.05 i 1.06 Reach minus Gier 17.03 :‘KL09 * -.87 :‘1.27 88/85 plus Everett minus Geir plus Reach -.95 i 6.39 1.14 1 .84 * Neither the first nor the second contrast is significant; however, the third, which compares the Big Brother agencies with the school agencies, is significant, in that volunteers affiliated with Big Brother agencies may stay as much as two months longer with the agency. A second main effect due to agency is significant for the March supervisor evaluations (p < .002). The agencies differed in two contrasts, also summarized in Table 11 above. Inspection of the means in Table 12 shows that the ratings were highest in Everett and Reach, fol- lowed by Big Brother/Big Sister, then Gier Park. Table 12. AGENCY EFFECTS: Combined Cell Means and Standard Deviations of Criterion Measures L_. Variable BB/BS Everett Reach Gier Park VOLUNTEER MEASURES Activity Satis. X 43.97 39.09 41.94 43.11 SD (5.42) (8.40) (7.399) (12.9) N 30 ll 16 9 Supervis. Satis. - X 47.93 ‘ 43.91 44.5 43.67 SD (5.83) (8.99) (5.46) (12.87) N 30 11 16 9 Months X 7.31 6.38 5.31 5.08 SD (0.52) (1.51) (2.48) (2.22) N 33 13 19 12 Teacher Satis. X 44.31 43.37 SD (6.12) (15.28) N 16 8 Role Clarity X 25.2 23.36 22.25 25.89 SD (3.08) (3.74) (5.94) (11.03) N 30 ll 16 9 Teacher Eval. Fall X 42.94 33.7 SD (7.24) (18.27) N 17 10 Teacher Eval. Winter X 43.67 40.43 SD (7.11) (7.61) N 12 7 Super. Eval. Winter X 31.44 41.82 45.33 26.29 SD (12.05) (3.55) (3.41) (10.97) . N 32 11 12 7 PROCESS MEASURES Vol. Interaction X 49.81 47.17 51.13 49.62 SD (4.9) (4.5) (4.49) (2.3) Child IHteraction 32 12 15 8 X 47.72 44.83 46.6 46.88 SD (5.79) (5.53) (6.33) (3.17) N 32 12 15 8 31 Table 12r-Continued Variable BB/BS Everett .Reach Gier Park CHILD MEASURES GPA Spring X 2.62 2.25 2.23 2.09 SD (1.05) (.78) ( 68) (.84) F‘ N 16 12 17 11 Absent X 2.59 3.19 5.98 2.39 SD (1.23) (8.01) (6.92) - (.95) N 16 12 17 ll Tardy ' X 40.31 35.42 32.94 46.82 , SD (33.83) (31.29) (18.06) (31.71) g_ N 16 12 17 ll Self-Concept X 3.5 3.86 4.46 4.19 SD (2.48) (2.64) (1.42) (2.33) N 25 13 18 11 Trust (w/covariate) 5.27 5.3 4.73 4.78 SD (.84) (0.55) (.89) (1.27) N 15 10 15 9 32 Hypotheses four and five were analyzed by way of correlations and are reported in Tables 13 and 15, respectively. Table 13. Correlations Between Meeting Goals and the Criterion Measuresa Met Goals Met Goals Variables Fall Winter N = 14 N = 28 Activity Satis. .44 ' .31 Supervis. Satis. .35 .03 Months -.18 .23 Teacher Satis. .16 .21 Supervis. Eval. Fall -.48 Supervis. Eval. Winter .45 ** Teacher Eval. Fall .05 Teacher Eval. Winter -.55 ** Vol. Interaction .30 .30 Child Interaction .46 .42 * Trust .05 .10 Self-Concept .28 .40 * GPA Spring .20 .52 ** Absent .29 -.10 Tardy .26 .13 a Sign on self-concept is reversed so that a positive number indicates a high degree of self-concept. p < .05 * *p<.01 The fourth hypothesis, concerning the degree to which contracted volunteers met their goals, in relationship to their satisfaction and effectiveness, is not significant for any of the criterion measures in the fall. However, the correlations suggest that a volunteer who meets his goals in the winter tends. to have higher supervisor evaluations, greater perception of his interaction with his child, and his child has a higher self-cancept.and grade point average. With respect to the 33 latter, these results should be interpreted somewhat cautiously, since it might be that a child with a higher self-concept and grade point average might aid the volunteer in meeting his goals. Also, the teacher's evaluation negatively correlates with the vol- unteer attaining his goals, the oppositeeme=m mo mezpwemwm emmmezpo> mo meepwemwm mexvp pemzoep mmewe Teoeeme m_eweo>ww umoz mmewseoeeme nmpmmexm ewe» meoz mmeweeoeeme eo _m>m~ nmmumexm muewEeoeeme nmpmmexm ewe“ mmmm bee: memzoep muewEeoeeme mpeweo>ween pmoz mmemsewwuuw nmmmmexm en ~m>m4 meom,m.emmpe:~o> mpwom mimemdq hu I < x~ozmmm< 42 APPENDIX B - VOLUNTEER QUESTIONNAIRE Code We are trying to find out what volunteer characteristics work best with what type of kid or adult. During the year, we will be asking you to fill out questionnaires periodically, and will relate this to how your kids do and perceive your relationship. This information will be used only for research and will NOT be used to evaluate you. We greatly ap- preciate your cooperation in sharing this information with us and it will be kept confidential. 1. Name: Student Number last first middle 2. Address: number and street city, state and zip code phone number 3. Sex: male female 4. Birthdate ,/ gg/ 5. Race: 6. What is the size of the place where you lived Caucasian when you went to High School? Negro Farm (3 acres or more) Mexican Rural area or town Chinese less than 10,000 Indian Small city (10,000 to 50,000) Japanese Large city (50,000 Other (Specify) to 125,000) City over 125,000 no information I {I ' I l | .l II III I I‘ll I II I II II III. 43 7. Year in College: 8. First term at MSU? __yes __no Freshman 9. College major Sophomore 10. College minor Junior 11. College GPA Senior 12. High School GPA Graduate student ____0ther (specify) 13. How man children did your parents have? (Circle the appropriate number.) 1 2 3 4 5 6 7 8 9 10 14. What was the order of your birth in the family? That is, were you the oldest, next oldest and so on? (Circle the appropriate number Showing order of birth from the oldest to the youngest.) l 2 3 4 5 6 7 8 9 10 15. What did your father (or father substitute) do for a living while you were growing up? 16. What was your mother's (or mother substitute) occupation? 17. What was the last year of school- 18. What was the last year of ing your father completed? schooling your mother completed? Graduate professional train- ing Graduate professional training College/University gradua- tion College/University graduation Partial college Partial college High school graduate . High school graduate Partial high school (10 or 11 Partial high school (10 or 11) Junior high school (8 or 9) _ Junior high school (8 or 9) 44 17. Continued 18. continued Seven years or less of Seven years or less of school school 19. What is your present marital 20. How often do you attend status? church? Never married Once or more weekly Married Not more than 3 times monthly Separated . Seldom (e.g. on special Divorced occasions) Widowed Never 21. What is your attitude toward 22. When you were in high school religion? By this I mean you how many extracurricular ac- are...' tivities did you participate in? Extremely devout Devout 23. During college how many extra- currivular activities have Moderate you participated in? Lukewarm 24. Have you held any elective or Not at all appointed offices in school, church, club, or any organized group during high school or college? ___yes ___no 25. Have you ever worked with young people before? yes ___no If yes, please briefly describe what you did. 26. Please list any previous volunteer work you have participated in. 45 27. Check the reason(s) why you are volunteering. _____to meet other volunteers ____to help others ____ppens up job possibilities ____to get dates _____it looks good on my academic record ____jt will help me in my degree program ____pther volunteers say it is a good/fun experience I know a kid/adult who had a volunteer, and it seemed like a good thing. I volunteered before and liked it. it supplements my academic record is a requirement for my degree program Other, please specify 46 APPENDIX C - VOLUNTEER EVALUATION Volunteer's Name Date Rater's Name Code As part of the ongoing MSU volunteer research project, we are con- cerned with your judgment of the volunteer's effectiveness. Please try to describe your perception of the volunteer you work with on the following items. 00 not Skip items; when you are unsure, give us your best guess. This information will be used only for research and will not be shown to the volunteer. We greatly appreciate your cooperation Tfi—Sharing_this information with us and it will be kept confidential. ITEMS TO BE RATED 21. Plans activities/ 4 3 2 l 0 Does not plan activi- assignments ties/assignments 22. Interacts well with 4 3 2 l 0 Interacts poorly with children children 23. Completes assignments 4 3 2 1 0 Does not complete assign- ments 24. Attends regularly 4 3 2 l 0 Is frequently absent 25. Initiates contact with 4 3 2 1 0 Does not initiate con- me tact with me 26. Shows resourcefulness 4 3 2 1 0 Lacks resourcefulness in in helping provide en- helping provide enrichment richment experiences experiences for child 27. Is punctual 4 3 2 1 0 Is often late 28. Asks for my advice 4 3 2 1 0 Does not ask my advice 29. Is able to accept 4 3 2 1 0 Unable to accept criticism criticism 30. Keeps me informed of 4 3 2 l 0 Does not keep me informed child's progress of child's progress 31. Arranges extra time to 4 3 2 l 0 Spends "required" time meet with child with child ‘ 32. Puts in additional time 4 3 2 l 0 Does not put in additional if situation arises time if situation arises 33. Excellent overall 4 3 2 l 0 ‘ Poor overall performance performance 34. Would you like to continue working with this volunteer? yes ___no 47 m n m N F .w m n m N F .m m n m N F .N m n m N F .F meow emmpenFo> m n m N F .N m e m N F .o m n m N F .m m e m N F .e m w m N F .m m n m N F .N m e m N F .F Fwow Amemm< . m n m N F .29.: 2955 :3: muewEeOFeme muewEeOFeme nmpmmexm mmewEeOFeme Fo mueweeoFeme menace“ muewEeoFeme mFeweo>wF “mos ewep meoE Fm>mF nmummexm nmpomexm ewe“ mmmF mFeweo>ere “mos .3oFme memm meg meFme .Fwom Fweomema new Amemmw emwm eOF muewEeoFeme Fo Fm>mF m .emmpeeFo> mew mmeFemmmn mmme uwem mmeoemme mew mFmeFo mmmee Fmoemm mmmmoome mmszaeo> I a xFozmme< BIBLIOGRAPHY BIBLIOGRAPHY Books Campbell, D.T., and Stanley. Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNally & Company, 1963. Eysenck, H.J. Eysenck Personality Inventory, California: 'Educational and Industrial Testing Service, 1968. Goodman, Gerald. Companionship Therapy. California: Jossey-Bass, 1972. Mager, Robert F. Preparing Instructional Objectives. California: Fearon Publishers, 1962. Mager, Robert F. Goal Analysis. California: Fearon Publishers, 1972. Mc Daniel, E.L. A Manual for the Inferred Self-Concept Scale. Texas: 1969, reprinted 1970. Mitchell, W.E. “The use of College Student Volunteers in the Out- patient treatment of troubled children" in Mental Health with Limited Resources. ed: H.R. Huessy. Newlbrk: Grune and Stratton, 1966. 28-37. ‘ 0diorne, G.S. Management by Objectives. New York: Pitman, 1965. Schindler, Eva and Rainman. The Volunteer Community--creative use of human resources. Washington: NTL Learning Resources, 1971. Smith, Patricia Cain. Measurement of Satisfaction in Work and Retire- ment, A Strategy for Study of Attitudes. Chicago: Rand McNally, 1969. Sobey, Francine. The Nonprofessional Revolution in Mental Health. New York: Columbia University Press, 1970. Truax, Charles and Mitchell, K. "Research on Certain Therapist Inter- personal Skills in Relation to Process and Outcome" in Handbook of Psychotherapy and Behavior Change. ed: Bergin and Garfield. John Wiley & Sons, Inc., 1971. ' 48 49 Journals Bryan, Judith F. and Edwin A. Locke. ”Goal-Setting as a Means of Increasing Motivation," Journal of Applied Psychology, No. 51, 274-277. . Cauley, John. "Job Descriptions: Matching Volunteer Skills with Agency Needs," Synergist, 2 (Winter 1974), 36~42. Duchastel, Philippe C. and Merril, F. ("The effects of Behavioral Objectives on Learning: A review of empirical studies," Review of Educational Research, 43, 1973. Gruver, Gary. "College Students as Therapeutic Agents," Psychology Bulletin, Vol. 76, No. 2 (1971), 111- 127. Goal Attainment Scaling Workshop, ed: Garwick and Vanderpool. Minnesota: 1973. Hogan, Robert. "Development of an Empathy Scale," Journal of Consulting and Clinical Psychology, 33 (1969), 306-307. Johnson, Thomas. "Role Ambiguity, Role Conflict, and Satisfaction: Moderating Effects of Individual Differences," Journal of Applied Psychology, Vol. 60, No. 3, 329- 333. Kiresuk, Thomas. "Goal Attainment Scaling: A General Method for Evaluating Comprehensive Community Mental Health Programs," Com- munity Mental Health Journal, 4 (1968). ' Lawler, E. "Job Attitudes and Employee Motivation: Theory, Research and Practice," Personnel Psychology, 23 (1970), 223-238. Locke, E.A. "Relationship of Intentions to Level of Performance," Journal of Applied Psychology, Vol. 50 (1966), 60—66. Locke, E.A. "Toward a Theory of Task Motivation and Incentives," Organizational Behavior and Human Performance, Vol. 3 (1968), 157-189. Locke, Edwin A. "Perfonnance Goals as Detenninants of Level of Per- fgrmance and Boredom," Journal of Applied Psychology, Vol. 51 (1967), 0-130. Locke, E.A. and Bryan, J. "The Directing Function of Goals in Task Performance," Organizational Behavior and Human Perfonnance, Vol. 4 (1969), 35-42. Locke, E.A., Cartledge, N. and Knerr, C.’ "Studies of the Relationship Between Satisfaction, Goal-Setting, and Perfonnance," Organiza- tional Behavior and Human Perfonmance, Vol. 5 (1970), 135-158. 50 Lyons, Thomas F. "Role Clarity, Need for Clarity, Satisfaction, Ten- sions and Withdrawal," Organizational Behavior and Human Perfor- mance, Vol. 6 (1971), 99-110. Rizzo, J.R., House, R. and Lirtzman. "Role Conflict and Ambiguity in Complex Organization," Administrative Science Quarterly, Vol. 15 (1970), 150-163. Sorum, J. "A Guide to Managing Student Volunteer Programs: An Objective- Centered Approach," Synergist, Vol. 3 (Spring 1974). Yelon, Stephen L. and Schmidt, William H. "The Effect of Objectives and Instructions on the Learning of a Complex Cognitive Task," The Jour- nal of Experimental Education, Vol. 41 (Spring 1973), No. 3, 91-95. Government Publications Wisconsin Bureau of Census. Methodology and scores of socioeconomic status paper #15, 1963. MICHIGAN STATE UNIV. LIBRQRIES llIlllllll|||||||IIIllllllillIlillllllllllllllllllllllllllllll 31293105502169