Teaching Moderated Multiple Regression for the Analysis of Mixed Factorial Designs

Traditionally, researchers have analyzed mixed factorial designs by dichotomizing the continuous variable and performing a factorial analysis of variance. Several problems surround the use of this approach, prompting many statisticians to recommend using moderated multiple regression as an alternative. This article presents the case for moderated multiple regression and provides on overview of the procedure.

Download Full-text

Programs for Problems Created by Continuous Variable Distributions in Moderated Multiple Regression

Organizational Research Methods ◽

10.1177/1094428106286984 ◽

2006 ◽

Vol 9 (4) ◽

pp. 554-567 ◽

Cited By ~ 16

Author(s):

Brian P. O’Connor

Keyword(s):

Multiple Regression ◽

Continuous Variable ◽

Moderated Multiple Regression

Download Full-text

Remarks on Scoring Androgyny as a Single Continuous Variable

Psychological Reports ◽

10.2466/pr0.1981.49.3.887 ◽

1981 ◽

Vol 49 (3) ◽

pp. 887-890 ◽

Cited By ~ 8

Author(s):

Robert F. Strahan

Keyword(s):

Multiple Regression ◽

Analysis Of Variance ◽

Continuous Variable ◽

Single Variable ◽

Psychological Androgyny ◽

Masculinity And Femininity ◽

Continuous Measures

Properties of two newly proposed, continuous measures of psychological androgyny are discussed in relation to an androgyny-relevant measure previously described. The difficulties of reflecting the construct of androgyny in a single variable are noted, and a combination of masculinity and femininity measures, through analysis of variance or multiple regression procedures, is recommended.

Download Full-text

IMBALAN DAN GAYA KEPEMIMPINAN PENGARUHNYA TERHADAP KEPUASAN KERJA KARYAWAN DI BALAI BESAR INDUSTRI HASIL PERTANIAN BOGOR

Jurnal Ilmiah Binaniaga ◽

10.33062/jib.v1i01.146 ◽

2018 ◽

Vol 1 (01) ◽

pp. 17

Author(s):

Ramlan Ruvendi

Keyword(s):

Job Satisfaction ◽

Multiple Regression ◽

Correlation Coefficient ◽

Analysis Of Variance ◽

Partial Correlation ◽

Multiple Correlation Coefficient ◽

Multiple Correlation ◽

Average Value ◽

Partial Correlation Coefficient

The study was carried out to find out whether there were influence and correlation bet-ween : a) Reward received by the IRDABI’s employees on their job satisfaction. b) style of the leader-ship on the job satisfaction. c) Reward together with style of leadership on the job satisfaction of IR-DABI’s employees.The result of the study showed that there was significant correlation and influence between the reward on the job satisfaction with was shown by the value of partial correlation coefficient of 0.6185 and coefficient of multiple regression for reward variable (β1) of 0.412. The influence of variable for style of leadership on the job satisfaction was also significant with the partial correlation coefficient of 0.5495 and coefficient of multiple regression (β2) of 0.355.In the test of Analysis of Variance (ANOVA) on the equation of multiple regression show that F-value was bigger that F-table (F = 58.97 > F-table = 3.098) or the Probability Value smaller than 0.05. At showed that there was significant correlation and influence between reward variables all together with style of leadership on the job satisfaction of employees. The value of multiple correlation coefficient (R) was 0.751 and R Square (R2) was 0.564. Value of R Square (0.564) meant that 56.5% of variation pro-portion total of job satisfaction can be eliminated of equation of multiple regression was used as the es-timator rather than using average value of job satisfaction as the estimator.

Download Full-text