Strategies for the selection of log-linear models

In a multidimensional contingency table strategies have been proposed to build log-linear models using either stepwise methods or standardized estimates of the parameters of the saturated model. Brown (1976) proposed a two-step procedure to screen effects and then test a subset of models. Alternate methods of model building are discussed with respect to the final choice of model and with respect to intermediate information available to the data analyst during the selection process.     

Nonstandard log-linear models.

This article demonstrates several useful varieties of nonstandard log-linear models. Some can be derived as nonhierarchical models by deleting lower-order effects in hierarchical models, but most often they will arise as the result of special hypotheses that the researcher wants to test. Three approaches to testing nonstandard models, partitioning chi-square, creating homogeneous subgroups, and the model matrix approach, are illustrated on a variety of data sets. It is assumed that the reader has been exposed to the general area of log-linear models, is familiar with dummy and effect coding of categorical variables in the usual regression framework, and knows how to express models in matrix form. A brief summary of the standard approach to log-linear models and of coding methods is provided for readers needing a review. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Application scenarios for nonstandard log-linear models.

In this article, the authors have 2 aims. First, hierarchical, nonhierarchical, and nonstandard log-linear models are defined. Second, application scenarios are presented for nonhierarchical and nonstandard models, with illustrations of where these scenarios can occur. Parameters can be interpreted in regard to their formal meaning and in regard to their magnitude. The interpretation of the meaning of parameters is the main focus of this article. Design matrices are used to describe the hypotheses tested in models and to illustrate cases in which parameters are interpretable. Also, design matrices are used to show where and how nonstandard models differ from standard hierarchical models. Coding schemes are discussed, in particular, dummy coding and effects coding. Data examples are given with data and models discussed in the literature. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Internship selection and early acceptance procedures: Students' evaluation.

We surveyed 192 graduate students in clinical psychology for the effects on predoctoral internship selection strategy of early acceptance procedures. Results showed a generally negative emotional response with strategy changes by students toward greater conservatism. We identified a subgroup that seemed to experience these effects more intensely. Implications of these findings for future selection procedures are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Use of log-linear models for assessing differential item functioning in a measure of psychological functioning.

Log-linear models are used to investigate contingency tables that cross-classify respondents according to item response, mental health status (MHS), and the background variables of ethnicity and gender. Specifically, log-linear models are used to examine item validity, defined as an item response by MHS interaction, and differential item functioning (DIF), defined as an interaction between item response and a background variable. The investigation focused on a set of items that measure subjective well-being and coping behavior. Female (n?=?627) and male (n?=?338) respondents represented 3 ethnic groups: African American, Anglo-American, and Hispanic/Latino. Strong evidence of item validity and some evidence of DIF was found. Most of the interaction between item response and either ethnicity or gender occurred among Ss with diminished mental health. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

On manuscript evaluation procedures.

Comments that the article by D. D. Bowen et al (see record 1973-11147-001) expresses dissatisfaction with the reliability and objectivity of referees. Other studies (e.g., J. Bernard, 1964) suggest that women are considered less seriously than men in intellectual realms. The author proposes evaluation procedures that assure anonymity of author and institution as one way of avoiding bias, and suggests that the American Psychological Association (APA) revise its bylaws to make this the policy for all APA journals. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Conceptual issues in the development of police selection procedures.

Although police selection procedures have been identified as necessary to the development of socially effective and responsive police agencies, the implication that such procedures are sufficient to this goal is criticized in the present article. It is argued that job performance criteria, embedded in the status quo of the prevailing police culture, inherently limit the ultimate utility of selection procedures. Therefore, prior to developing selection and job performance prediction procedures, which are at best palliative solutions to the problems of the police agency, it is necessary to address the more fundamental issues regarding social goals of a contemporary police agency. Such an analysis may suggest alternative ways of providing the services necessary to obtain these identified social goals. (35 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

An evaluation of modeling and experiential procedures for self-disclosure training.

Attempted (a) to adapt and implement an interpersonal-skills training paradigm for training married individuals to exhibit increased self-disclosive behavior in their marital relationships, and (b) to determine the efficacy of experiential and modeling procedures in such training. 27 university-affiliated married couples (mean age 24.5 yrs, mean length of marriage 3.1 yrs) were randomly assigned to 1 of 3 groups: a communication-skills training group, a modeling group, and a control group. Frequency, duration, and quality of self-disclosive behavior were analyzed for pre- to posttraining changes by means of the Hill Interaction Matrix and the Facilitative Self-Disclosure Scale. Results indicate that (a) the communication-skills training group demonstrated a significant increase on 4 of the 5 self-disclosure measures, (b) the modeling group demonstrated an increase on 3 of the self-disclosure variables, and (c) the controls did not change on any of the variables. Implications of integrating both modeling and experiential methods into interpersonal-skills training programs are discussed. (29 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Impact of valid selection procedures on work-force productivity.

Used decision theoretic equations to estimate the impact of the Programmer Aptitude Test (PAT) on productivity if used to select new computer programers for 1 yr in the federal government and the national economy. A newly developed technique was used to estimate the standard deviation of the dollar value of employee job performance, which in the past has been the most difficult and expensive item of required information. For the federal government and the US economy separately, results are presented for different selection ratios and for different assumed values for the validity of previously used selection procedures. The impact of the PAT on programmer productivity was substantial for all combinations of assumptions. Results support the conclusion that hundreds of millions of dollars in increased productivity could be realized by increasing the validity of selection decisions in this occupation. Similarities between computer programers and other occupations are discussed. It is concluded that the impact of valid selection procedures on work-force productivity is considerably greater than most personnel psychologists have believed. (37 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Three models of evaluation.

There are at least three schools or types of evaluative research: Evaluation by Objectives, a Systems Approach and a model proposed by Ackoff et al (1962). In the Evaluation by Objectives approach, a program objective is selected and efforts are made to determine the degree to which the objective is achieved. The Systems Approach is less concerned with a determination of the present level of functioning and is more concerned about increasing the level of achievement. According to Ackoff the goal of evaluative research is the selection of the "best" from amongst alternate ways to achieve a desired goal. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Lessons from arithmetic: Influences on the selection of problem-solving procedures.

Mental arithmetic involves an intriguing interaction between (a) memory and problem-solving processes that are domain independent and (b) solution procedures and factual knowledge that are specific to the domain of arithmetic. Relations between domain-independent and domain-specific knowledge are of central concern to many developmental, cognitive, and instructional psychologists, and consequently the performance of children and adults on mental arithmetic tasks have come under increasing scrutiny. Despite this common interest, in recent years the focus of research on adults has diverged somewhat from that of research on children. Recent research with children has concentrated on the selection process, the process by which one of several available procedures is selected for use on specific problems. In contrast, it is commonly assumed that adults always or nearly always use fact retrieval to answer simple arithmetic problems. Thus research with adults has been focused primarily on processes and knowledge structures involved in the retrieval of arithmetic facts, and the selection process has been virtually ignored. This article describes two studies which help to highlight the value of focusing on the selection process as a common and dynamic feature in the performance of children and adults. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Selection of pairwise multiple comparison procedures for parametric and nonparametric analysis of variance models.

Discusses 2 issues that are often overlooked in the evaluation of pairwise multiple comparison procedures (MCPs): Comparisons of powers for competing MCPs are not useful unless the procedures have identical nominal experimentwise error rates; and "protected" MCPs do not control the experimentwise Type I error rate at the nominal alpha except in the complete null case. Also, procedures based on the ranks of all K groups do not test the same hypothesis as MCPs that involve reranking the data for each pairwise comparison. (41 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The contribution of interview and situational performance procedures to the selection of supervisory personnel.

2 groups of supervisors (each N = 40), selected to be as similar as possible on a set of control predictor variables "known to be predictive of supervisory performance" (basic abilities test score, supervisory practices score, age, service, job level) and to be maximally different on criterion scores of supervisory effectiveness ratings, were used to try out 5 experimental supervisory performance predictors: panel interview, individual interview, group discussion problem, role playing situation, and small-job management. Each group was divided into a high and low subgroup on the basis of the control predictor variables. The predictor battery validity increased from an average of .27 for the control predictors alone to .30-.34 when various of the experimental predictors were added. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A method for empirically assessing volunteer selection effects: Recruitment procedures and responses to erotica.

Examined whether 58 male and 47 female undergraduates who were recruited for a study on erotica responded in a similar manner to an erotic film as 29 male and 26 female undergraduates who were recruited for research on a neutral topic, personality questionnaires. In Part 1 of the study, all Ss answered questionnaires, including the Bem Sex-Role Inventory and a sexual opinion survey, that assessed personality and demographic characteristics. In Part 2, Ss watched the erotic movie and completed a feelings scale. Findings indicate that conclusions about prior sexual experience, sexual opinions, and affective responses to the erotic film were qualified by recruitment condition. Ss who volunteered for the experiment on erotica were more sexually experienced and, for women, were significantly more erotophilic than Ss who volunteered for the research on personality questionnaires. Results suggest that volunteer selection can have an important effect on experimental results and demonstrate a method for assessing the extent of these effects. (33 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Interpersonal skills and pretraining: Implications for the use of group procedures for interpersonal learning and for the selection of nonprofessional mental health workers.

Used a behavioral assessment procedure to rate 138 undergraduates as high or low in interpersonal skills. After receiving practice, cognitive, or no pretraining in group procedures Ss participated in discussion groups. Groups composed of members who were rated high in interpersonal skills were found to engage in significantly more personal discussion and feedback and less impersonal discussion than groups composed of members rated low in these skills. Groups receiving pretraining showed similar significant differences on these dimensions when compared to no-pretraining controls. Of the 2 types of pretraining investigated, the cognitive approach with no practice trials appeared most effective. Group composition and pretraining interacted such that groups composed of high-skill participants were affected by specific pretraining conditions, whereas low-skill groups were generally not differentially responsive. Implications of these findings for the most efficient use of group procedures and selection of nonprofessionals are discussed. (26 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Interviewer consistency in the use of empathic models in personnel selection.

Personnel selection interviewers predicted the responses of applicants to 2 paper and pencil tests. Comparisons were made between these predictions and the responses of applicants (accuracy score), predictions and the responses of interviewers (similarity score). (These scores were referred to collectively as empathy scores.) Statistical analysis of empathy scores and their components indicated marked inter-interviewer inconsistency, which was interpreted to mean that interviewers tend to make errors by resorting to empathy as a basis of decision making. Explicit, actuarial bases of decision making in interviewing are advocated. Results of a follow-up study supported this interpretation. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Elasto-optic evaluation of stabile models. Stress concentration

The nutritional quality of rice protein was compared with that of whole egg protein by slope ratio assay. Diets for each food at four levels of protein, 4, 6, 10 and 15% and a protein-free diet were given to male weanling rats of the Sprague-Dawley strain for 21 days. The slopes of the regression lines of the whole egg and rice groups calculated from the changes of body weight (Y in g/21 days) with nitrogen intake (X in g/21 days), including and (excluding) zero protein group were, respectively, Y=27.39 X-12.26 (Y=24.41 X-1.86) and Y = 13.86 X-8.06 (Y = 12.54 X +0.50). Assuming a potency of 100 for the egg protein, the relative potency of rice estimated from body weight gain with nitrogen intake was 51 (51). The values for rice calculated from body water gain and nitrogen retention with nitrogen intake were, respectively, 51 (47) and 46 (44). These values were compared with RNV of several varieties of conventional rice and high-protein rice.