Performance evaluations were obtained on firefighters in a large Midwestern City. The evaluations were conducted through utilization of two different scale types (The Behavioral Expectation Scale and The Mixed Standard Scale). These evaluations were obtained in order to test the hypotheses that the MSS was psychometrically superior to the BES in the reduction of halo and leniency error and that the MSS was also the better scale type in terms of producing higher interrater reliability. Leniency error (in both the absolute and comparative sense) was examined by conducting a series of T-tests. Halo error was investigated by a comparison of the means of the dimensions from each scale. The technique used to assess interrater reliability involved estimating the reliability of the differences in the shape and level of performance profiles of firefighters. The results showed that the first hypothesis, which proposed that use of the MSS produced less halo and leniency error than did use of the BES, was supported; however, the second hypothesis, which proposed that use of the MSS produces higher levels of interrater reliability, was not supported.

A review of the literature on decision making indicated a lack of research attention given to the effects of the dynamics of the social context surrounding participation on decision outcomes. The present study examined the relationship between social context and group decision making formats, and its implications for the effective implementation of group participation in decision making The effectiveness of three decision -making formats, the nominal technique, the interacting technique, and the consensus technique, were compared under conditions of cooperation and competition on the three decision effectiveness dimensions of quality, acceptance, and synergy. Two-hundred and forty-eight undergraduate students, working in four or five-person groups, solved the NASA Moon Survival Problem under one of six conditions. Quality, synergy, and both self -report and behavioral measures of acceptance were assessed. An 3 X 2 (decision format X social context) analysis of variance indicated that social context and decision format have no effect on measures of decision quality, behavioral acceptance, and synergy, although the results were in the hypothesized direction. The study did indicate that self-reported acceptance, satisfaction with the dynamics within participation, and representation in discussion were contingent upon the social context. Cooperative groups facilitate the acceptance of the groups' decision and were more satisfied with the dynamics within participation. Also, competitive groups felt that people did not dominate the discussion in their groups more than cooperative groups.

An Army training program was evaluated in order to present recommendations for program refinement. The training program, developed as part of two larger Army combat development (CD) studies (DuBois & Smith, 1989a, 1989b), represents an attempt to define program design requirements for training tank commanders to use the Intervehicular Information System (IVIS). IVIS is a computer-based command, control, and communications system proposed for future M1A1 tank upgrades. Twenty-four Army tank commanders (TCs) participated in the research effort. TCs were of various rank and age. Training and testing were conducted over a 4 day period. Training lasted 1-1/2 days followed by 2-1/2 days of testing. The Biographical Questionnaire, the IVIS Knowledge Test, the IVIS Performance Test, the SIMNET Knowledge Test, the IVIS Survey, and the IVIS Training Reactions Questionnaire were administered to each TC. TC biographical data, reactions to training, post-training performance, and post-operational performance were evaluated. Scale reliability was evaluated for each IVIS instrument. Descriptive statistics were calculated for the soldier reaction and performance measures. Regression analyses were performed to assess the impact of TC background characteristics on commander performance. Results of the above analyses indicate: (a) the IVIS Performance and Knowledge tests are reliable, (b) the soldiers learn during training and the learning transfers to the operational setting (performance measured after two days of testing shows soldiers performance significantly increased from that of post-training performance), and (c) soldier biographical characteristics do not predict IVIS proficiency. The results call for more training time allotted to teaching IVIS skills with some IVIS functions deserving more attention. Additionally, it appears personnel requirements may not be an issue for selecting IVIS system users as the study found no correlation between biographical characteristics and IVIS proficiency. And finally, it is recommended that observational raters receive greater rater training as interrater reliability estimates were low.

Despite the controversy surrounding Affirmative Action (AA), relatively little research has appeared about attitudes towards these programs. In this research, an exploratory approach is implemented to assess the support of personnel managers for the theory of AA as well as the mechanisms designed to carry it out. Also, the relationship to Supreme Court decisions, relevant legislation, and numerous demographic, attitudinal, and organizational variables are examined for their impact on Affirmative Action attitudes. A questionnaire was utilized to assess support for AA and its correlates. The majority of personnel managers indicate support for both AA in theory and the mechanisms required to carry these programs out. This research indicates either very small or no differences exist between support for AA concepts versus AA mechanisms, support for gender -based versus race-based AA, or support reported by private sector versus public sector personnel managers for AA. How personnel managers perceive the impact of Court Decisions and the 1991 Civil Rights Act on AA implementation, although small, was found to be a significant correlate of AA attitudes. The race of the respondent was found to be the most significant determinant of AA attitudes. Although minorities were found to be more supportive than nonminorities, both were found to register support for these programs.

The purpose of this study was to develop and validate a Weighted Application Blank to predict turnover for use as a selection tool for a large retail organization. Utilizing the England (1971) procedure, it was hypothesized that significant derivation and cross-validities would be obtained. The hypothesis was partially supported: the derivation validity coefficient was significant (r = .28, p < .01) and the resulting cross-validity coefficient was net significant (r = 19, p < .05). The results and recommendations for implementation of the WAB are discussed.

Research concerning the validity of various procedures for assessing managerial potential is reviewed. It is found that several assessment techniques are valid for identifying and/or predicting potential and that a combination of techniques usually accounts for a significantly greater amount of criterion variance than does a single assessment procedure used alone. The lack of research concerning the use of multiplecomponent assessment techniques for the selection of individual job applicants for lower-level managerial positions is pointed out. The present study concerns itself with an operational individual assessment program designed to identify and select from among experienced job applicants those individuals exhibiting greater supervisory and/or managerial potential. It's objectives are to examine the merits of this process from the organization's point of view and to assess the relative values of the separate components utilized in the process, namely clinical ratings and objective tests. The results tend to support the ratings, though no significant correlations were found. Weaknesses in the criterion are pointed out and it is suggested that both the assessment process and the performance appraisal procedure be revised.

When 32 elementary education students, 16 of whom were enrolled in an off-campus block of laboratory experiences with handicapped, took the Rosenzweig Picture-Frustration Study and the Kuder Occupational Interest Survey, Form DD, there was no evidence that the professional semester affected self-confidence or social adjustment as measured by the number of extrapunitive and need-persistence reactions, respectively, on the P-F Study. The t test was used to compare the mean number of those responses made by members of the two groups. There was considerable evidence, however, that the number of subjects who gave elementary education the same or an even higher rank following the professional semester was too great to be attributed to chance when compared by means of chi square to the number and direction of rank changes made by members of the control group on the same occupational survey. The support which this study has given to actual classroom experience as a cause of increased professional commitment is sufficiently great to imply that other students might profit if their programs contained periods of time spent in daily contact with learners before their student teaching experiences begin.

A study was conducted to develop a written tank knowledge test for United States Army armor trainees. Such a test may be employed as a tool for training program evaluation and as a criterion measure for assessing trainees. The Tank Knowledge Test (TKT) was developed from items written by subject matter experts familiar with the armor training program. After pilot testing, a revised version of the TKT had an internal consistency reliability of .79. Item difficulty levels ranged from .16 to .92 with a mean of .57. A factor analysis performed to examine the possibility of deriving subscales from four clusters of topically related items revealed no support for a four factor solution.

The theory of groupthink has been highly beneficial in the study of how groups make decisions. It has permeated almost every field containing decision making groups. Despite its popularity, there has been a surprising lack of empirical support for the model. It is the aim of this paper to suggest a possible explanation for the current state of groupthink research. First the groupthink model is described briefly, followed by a look at several selected empirical and case studies of groupthink. A potential reason for the dearth of empirical is then proposed along with a suggestion for future groupthink research.

Research has shown that more than 45% of American companies are opting to integrate personality tests in their recruitment processes. Given this surge in personality testing, this thesis examines whether personality testing is a valid predictor of job-fit and performance in the context of personnel selection. A large proportion of this paper is focused upon the Big-Five factor model, its limitations, and derivative tests of the model. The impact of technology upon personality testing is also discussed as an emerging field. By tracing and examining the history of personality testing to current day, I have found that personality tests are best administered when they provide incremental validity over other tools and are matched to specific job-criteria.