Social scientists are using Al (Artificial Intelligence) and Expert Systems in a variety of areas. This thesis explores some applications of "intelligent" programs in the field of sociology. These applications include modeling social networks using logic programming and simulating sociological decision-making using small knowledge-based systems. PROLOG and VP-Expert are used as development tools. In this thesis, PROLOG was used for modeling triads. In this situation, PROLOG was used to define the "actors," the relations between them, and rules governing the existence of transitive and intransitive triads. Users are able to query the knowledge-base about various aspects of this situation. In addition, VP-Expert was used to construct some small knowledge-based systems. These systems deal with determining whether a triad is transitive or intransitive, classifying adaptation to anomie (normative confusion), determining whether a type I or type II error has been made in the hypothesis testing process, determining a variable's level of measurement, and classifying individuals on the basis of socio-economic status.

It is one of the main contentions of this presentation that a discipline such as sociology must constantly refer to the socio-cultural circumstances within which it operates, in order to make decisions concerning the expansion or contraction of the area of phenomena it is attempting to analyze and study. Relevancy, admittedly, is a value based on an ideological assumption, but within the confines of this presentation remains a very important one; one which establishes both a direction toward which it is felt sociology must move as well as a base from which it must be extended.

Five-person continuing and zero-history groups solved sentence-completion problems in two communication networks: the comcon and the wheel. Oral communication was utilized and all problem solving sessions were tape recorded for analysis. Analysis of variance revealed that the comcon network was significantly faster that the wheel in completing the task. The continuing groups used significantly more messages than the zero-history groups in problem solving. A chi-square test indicated that the continuing groups also used significantly more "sociability" messages than the zero-history groups in solving the sentence-completion problems. Analysis of variance revealed that in both groups, subjects in the central position of the wheel network were more satisfied than the subjects in peripheral positions. No significant difference was found between continuing and zero-history groups in the amount of time required for problem solving.

During the social turmoil of the 1960's, the expression of anti-establishment attitudes and the consequent rebellion against authority brought law enforcement agencies across the United States into frequent and often violent confrontation with citizens. Responding to what was perceived as a negative image of police, law enforcement agencies began to change their traditional uniform to a "soft-look" civiliantype blazer uniform. The movement to change the uniform was based on the assumption--which had not been empirically examined--that the traditional uniform with its official insignia and other accouterments overwhelmed citizens and engendered a wide range of negative reactions and that removal of these symbols (i.e., the uniform) would result in a more positive relationship between the public and the police. The purpose of the present investigation was to assess the reactions of individuals to varying modes of police dress using a methodology that considered both physiological and cognitive components of affective responsivity. A 2 (race of participant) X 3 (style of police dress) between -within design was used. The between factor was race of the participant (i.e., black or white), and the within factor was style of police dress (i.e., full uniform, full uniform minus weapon and accessories, and "plain" clothes consisting of blazer, slacks, coat, and tie). Eighteen black and 18 white male undergraduate students viewed color slides that depicted the varying styles of police dress. During the first showing of the slides, heart rate and skin conductance were recorded. During a second showing, participants rated each stimulus slide using word descriptors in a semantic differential format. The results reflected that participants did not evaluate the various modes of police dress differently and that there were no racial differences in the evaluation of the various modes of police dress. The implication is that the cost of adopting the blazer-style uniform is not justified on the basis of reducing negative affective reactions toward police and that improvement in the public image of the police cannot be achieved by merely changing the uniform. It is suggested that the major benefit of the uniform change may be that it facilitates an examination by the individual police officer of his or her role in the community. Limitations of the present study and recommendations for future research are also discussed.

A review of the research on family caregivers of the frail elderly indicates that although caregivers often experience high levels of burden, they make only minimal use of available professional, or formal, services. A theoretical model of helping and coping proposed by Brickman, Rabinowitz, Karuza, Coates, Cohn, and Kidder (1982) suggests that attributions of responsibility for causing and for solving problems define four distinct attributional styles, each determining how people will respond to outside help. The current study of 40 family caregivers attempted to validate an attribution instrument based on the Brickman et al. (1982) model, and to determine whether scores from the instrument were predictive of formal service use. Four models were hypothesized to be identified through factor analysis, each coinciding with one of Brickman's attributional styles. It was also hypothesized that score totals for each of the attributional style models would be predictive of formal service use. Neither hypothesis was supported. However, evidence which did not reach statistical significance suggests the existence of the models in this group. Also, regression analysis found several demographic variables to be predictive of formal service use. These include; care receiver age, caregiver education level, and whether or not the caregiver lives with the care-receiver. Caution must be taken in generalizing the findings from this study due to the questionable validity of the measurement instruments and to the small sample size. Suggested future research includes further validation work on the assessment instruments, and the recruitment of a larger sample group.

The purposes of this study were (a) to determine if relaxation procedures, consisting of meditation and cognitive self-statements, were effective in improving test performance and in reducing test anxiety and (b) to investigate the relationship between test anxiety and test performance. Thirty-six students in an undergraduate psychology class were administered a self-report anxiety inventory, the Test Anxiety Scale-Revised (TASC-R). The students were then matched according to test anxiety level and randomly assigned to an experimental group or a control group. The experimental group received 20 minutes of relaxation treatments immediately prior to the third, fourth, and fifth course content exams. The control group spent an equal amount of time before these three course content exams in unstructured review. Students in both groups were asked to respond to the TASC-R prior to taking the initial relaxation treatment and following the final relaxation treatment. Analysis of variance procedures were used to assess the effects of the relaxation procedures on anxiety and test performance. The results showed no significant difference between the treatment and control groups on either anxiety or test performance. Thus, no evidence is presented in this study to indicate that relaxation procedures do in fact improve test performance, or reduce test anxiety. Correlational procedures were used to assess the relationship between test anxiety and test performance. No meaningful significant relationships were demonstrated for either group or both groups combined.

This study examined the relationship of cognitive complexity with urban-rural locality, communication apprehension, and world view. Emphasis was placed upon the population variable of urban-rural locality as it related to cognitive complexity; however, the variables of communication apprehension and world view were also examined in an attempt to discover an interrelationship among the variables. Using the Crockett Role Category Questionnaire, the short version of McCroskey's Personal Report of Communication Apprehension, a world view scale as developed by Dodd and Garmon, and selected demographic and group membership items, the researcher tested one hundred fifteen undergraduate college students. Data analyses included analyses of variance and correlation and regression analyses. Results of the study indicated a significant sex difference in which females were shown to have a higher degree of cognitive complexity than males. A multiple regression found the variables of number of children in the family, sex, and other organization membership to be chief predictors of cognitive complexity. The group membership variables were discovered to have a strong correlation with one another, with the other organization membership variable exerting the strongest influence. An analysis of variance, however, revealed the major finding of the study, a significant three-way interaction of the variables of population, world view, and communication apprehension with cognitive complexity.

Since 1884, the courts have embraced and applied the doctrine of employment-at-will which states that employers may terminate their employees, with or without cause or notice, at any time. Recently, the courts have recognized exceptions to this doctrine. In order to avoid these exceptions, lawyers and human resource specialists have suggested that employment materials should explicitly state that all employment is at-will. However, while stating an employment-at-will policy may protect employers' rights to terminate at-will, little is known about the consequences of adopting these measures with respect to applicant and employee attitudes. The concept of employment-at-will is explored and research that investigates the effects of explicitly stating these policies in corporate recruiting brochures is presented. First, the employment-at-will literature is reviewed. Second, variables are discussed that address the possible effects of stating an employment-at-will policy in corporate recruiting materials. Third, variables are presented that may interact with termination policy with regard to potential applicant attitudes. Fourth, a 2 X 2 X 2 factorial design was used to assess the influence of self-efficacy (high versus low), termination policy (terminate for "just cause" versus employment-at-will), and compensation policy (leading versus average) on future applicants' 1) favorableness of impression of the organization, 2) interest in pursuing employment opportunities, and 3) intention to join a union. indicated a interaction Results significant termination policy by self-efficacy for intent to pursue employment. Specifically, it was found that individuals with high self-efficacy were more favorable toward pursuing employment with an organization with a due process termination policy than an organization with an employment-at-will termination policy. Results also indicated that whether an individual has a union member in his family was associated with the potential applicant's ratings of corporate image (r = -.15, p < .05) and intent to pursue union membership (r = .23, p < .01). Finally, possible reasons for the current findings and avenues for future research are addressed.

One hundred and two fundamentalist Christians were administered scales measuring three religious orientations (intrinsic and extrinsic religion and religion-as-quest) and selective exposure. The aim of this study was to examine the relationships between the three orientations and selective exposure to religious information. Selective exposure was measured in three ways: Belief confirmation - seeking belief-supporting information; Selective avoidance - avoiding belief-contradicting information; and Differential exposure - seeking belief-confirming information while avoiding contradicting information. Results of this study indicate that both quest and intrinsic religion predict the desire to read belief-confirming information. For selective avoidance, only quest predicted willingness to read counter-attitudinal information, thus supporting Batson's contention that religion-as-quest uniquely predicts open-minded truthseeking and willingness to doubt. Differential exposure was primarily predicted by extrinsic religion.

This thesis investigates the effect of nine sociological variables on church attendance in two churches in Bowling Green, Kentucky. The sample was drawn from a Church of Christ (N=135) and a Southern Baptist church (N=105). The effects of these variables were determined by using the computer program "Multiple Nominal Scale Analysis." The explanatory variables examined with church attendance were other religious activity, education, occupation, income, meaning, belonging, conservatism/ orthodoxy, a general sense of meaning, and the number of other voluntary associations belonged to. Meaning was defined as the belief that one belonged to the New Testament Church. Belonging was divided into two variables; the number of closest friends that attended one's congregation and the number of closest friends in one's denomination. Eight other variables were examined for their effect on belonging, defined as the number of closest friends in one's congregation. They were how many years the subject has been attending his present church, how many years he has been a member of his present church, whether or not the subject's mother and father are living and whether or not each of them attended church, and if so where. Subjects were also asked how long they had lived in or near Bowling Green, Kentucky, and the occupations of one's five closest friends. The findings indicate that in both churches other religious activity was the most powerful predictor of church attendance. In the Church of Christ this was followed by belonging, defined as the number of closest friends in one's congregation, and meaning. In the Baptist church "social status" variables such as income, number of other voluntary organizations belonged to and occupation ranked below other religious activity. This supported the principal hypothesis that attendance in the Church of Christ would be higher and that meaning and belonging would be better predictors of church attendance here than in the Baptist church. Concerning belonging, the findings indicate that the occupations of one's closest friends is important in both churches but where the combined variable of number of years one has attended his present church and number of years he has been a member are important in the Baptist church, the background variable of the church one's father attended is more important in the Church of Christ.