Eugene F. Stone-Romero

Methodological Artifacts in Moderated Multiple Regression and Their Effects on Statistical Power

Monte Carlo simulations were conducted to examine the degree to which the statistical power of moderated multiple regression (MMR) to detect the effects of a dichotomous moderator variable was affected by the main and interactive effects of (a) predictor variable range restriction, (b) total sample size, (c) sample sizes for 2 moderator variablebased subgroups, (d) predictor variable intercorrelation, and (e) magnitude of the moderating effect. Results showed that the main and interactive influences of these variables may have profound effects on power. Thus, future attempts to detect moderating effects with MMR should consider the power implications of both the main and interactive effects of the variables assessed in the present study. Otherwise, even moderating effects of substantial magnitude may go undetected.

Relative power of moderated multiple regression and the comparison of subgroup correlation coefficients for detecting moderating effects.

A Monte Carlo simulation assessed the relative power of 2 techniques that are commonly used to test for moderating effects. The authors drew 500 samples from simulation-based populations for each of 81 conditions in a design that varied sample size, the reliabilities of 2 predictor variables (1 of which was the moderator variable), and the magnitude of the moderating effect. They tested the null hypothesis of no interaction effect by using moderated multiple regression (MMR). They then successively polychotomized each sample into 2, 3, 4, 6, and 8 subgroups and tested the equality of the subgroup-based correlation coefficients (SCC). Results showed MMR to be more powerful than the SCC strategy for virtually all of the 81 conditions

Type II Error Problems in the Use of Moderated Multiple Regression for the Detection of Moderating Effects of Dichotomous Variables

Monte Carlo simulation procedures were used to assess the power of moderated multiple regression (MMR) to detect the effects of a dichotomous moderator variable under conditions of: (1) between- group differences in within-group relationships between two variables (i.e., |ρ XY (1) -ρ XY (2) |= .20, .40, .60); (2) different combined sample sizes for the two groups (N 1 + N 2 = NT = 30, 60, 90, 180, 300); and (3) differing proportions of cases (P-i ) in the two groups (i.e., P 1 = .10, .30, .50). Results showed that, consistent with our a priori predictions, the power of MMR increased as: (1) total sample size (NT) increased; (2) the difference between within-group correlation coefficients increased; and ( 3) the difference between the proportion of cases in each group decreased. Moreover, the simulation showed that these three variables had interactive effects on power. The major implication of our findings is that in cases where tests of moderating effects are conducted with MMR and the proportion of cases in each group differs greatly, inferences of no moderating effect may be erroneous: Such inferences may be the result of low statistical power rather than the absence of a moderating effect.

Trends in Research Design and Data Analytic Strategies in Organizational Research

We coded all articles that were published in the Journal of Applied Psychology in the 1975-1993 period (N = 1,929) in terms of the type research design used in studies (e.g., experimental, nonexperimental), and the major strategies used in analyzing the data of the studies (e.g., analysis of variance, multiple correlation/regression, covariance structure analysis) and computed annual frequencies of use of the research designs and the data analysis strategies. These frequencies were used to compute yearly percentage use indices (PUIs) for each of the designs and data analytic strategies. Results showed that over the period considered: (I) there were fairly minor fluctuations in the PUIs for experimental and nonexperimental designs; (2) the PUI for covariance structure analysis procedures (e.g., LISREL, EQS) was very low during the 1975-1984 period but has increased dramatically thereafter; (3) the PUIs for multiple regression and several other multivariate data analytic strategies have increased somewhat over time; and (4) the PUIs for data analytic procedures that involve tests of mean differences have fluctuated over time, but there is a slight decrease in the use of ANOVA. Implications of these trends are considered.

Development of a multidimensional measure of perceived product quality

Abstract A measure of perceived product quality (i.e., the Perceived Product Quality Measure of PPQM) was developed that had scales for measuring four general dimensions of quality, i.e., flawlessness, durability, appearance, and distinctiveness. Results of two studies (preliminary and main) showed support for the reliability and validity of the PPQM. Results of analyses of data from the main, experimental study showed that: (1) the PPQM scales had fairly high levels of internal consistency; (2) manipulations of several attributes of products led to expected changes in the responses to PPQM scales; (3) the PPQM scales predicted a willingness to buy criterion; and (4) the results of a confirmatory factor analysis showed support for the hypothesized multidimensionality of the PPQM. In view of these findings, we believe that the PPQM should prove useful in future research on perceived product quality. It should also prove helpful in efforts to manage the quality of products produced by individuals on work organizations.

Estimating the Power to Detect Dichotomous Moderators with Moderated Multiple Regression

A QuickBASIC program for estimating the statistical power to detect the effects of dichotomous moderator variables using moderated multiple regression (MMR) is available. The program runs on IBM and IBM-compatible personal computers and estimates power based on specific values for (a) total sample size, (b) sample sizes across the two categories of the hypothesized moderator, and (c) correlation coefficients between predictor and criterion scores for each of the two moderator-based subgroups. The compiled run time and source code versions of the program can be obtained from the first author.

Cross-cultural differences in responses to feedback: Implications for individual, group, and organizational effectiveness

Appropriate (functional) responses to negative feedback are vital to both the short- and long-run performance of individuals in organizations, and, therefore, for their work groups and organizations. Regrettably, research shows that individuals may not respond appropriately to such feedback when they are motivated by self-enhancement. Moreover, it shows that self-enhancement tendencies are far more common among people in individualistic cultures than among people in collectivistic cultures. In view of this, we present a cross-cultural model of responses to feedback along with theory-based explanations for cultural differences in responses to feedback. In addition, we detail a number of strategies that egocentric people use for the purpose of either maintaining or enhancing their self-esteem when dealing with negative feedback. Finally, we offer suggestions for motivating people in individualistic cultures to respond to negative feedback in functional ways.

A Comparative Study of Hispanic‐American and Anglo‐American Cultural Values and Job Choice Preferences

Using data from 184 employed Hispanic‐American and Anglo‐American participants in the United States, the present study examined the relations between four cultural values (i.e., collectivism, power distance, familism, present time orientation) and job choice preferences. Results revealed that (1) collectivism was positively related to the importance of coworkers and working in a diverse organization, (2) familism was related to preferences for jobs with personal time off, and (3) power distance was related to the importance of organizational reputation and promotion opportunities. In addition, the findings revealed that, relative to Anglo‐Americans, Hispanic Americans felt that organizational reputation, flexible work hours, bonuses, and diversity were more important job choice factors. Implications are offered for conducting future research on job choice and developing recruitment practices in multicultural organizations in the United States.

3. The functional and dysfunctional consequences of human resource information technology for organizations and their employees

In recent years there has been a rise in the use of technology in the field of human resources (HR) management. For example, surveys show that 70% of large firms now use HR information systems, 80% conduct online recruiting, 67% post job openings online, and 40% use web-based portals as a means of communicating organizational policies and practices to actual and potential employees (Cedar, 2001; Towers Perrin, 2001). The primary reason for the increased use of human resources information technology (HRIT) is that organizations believe that automating labor intensive HR functions will help them streamline processes, utilize employees talents and skills more effectively,

Taking the strain out of negative affectivity : Development and initial validation of scores on a strain-free measure of negative affectivity

The purpose of the present study was to validate a measure of negative affectivity (NA) that does not contain strain content: The Strain-Free Negative Affectivity (SFNA) scale. A 33-item Likert-type scale was developed based on the negative emotional reactivity component of Watson and Clark’s constitutive definition of NA and was administered to two separate samples (ns = 225 and 281, respectively) of college students (ages 18 to 46), along with several extant measures of NA. Five items eventually were removed from the scale. Results from both studies support the construct validity of the SFNA scores, indicating support for the SFNA as a worthy alternative to extant measures of NA, especially in research examining stressor-strain relations.