I would like to measure manager effectiveness, and, in my survey, I collected 38 important items as independent variables as perceived from interviews (38 items).
What is the best way to use SPSS to find the outcome (effectiveness) ,e.g regression, collaboration, t-test, ANOVA ………….ETC. Thanks a lot
Regression most obviously is the best choice for you. Choose type by properties of the dependant variable. Google how to choose what regression. Best wishes David Booth
What is your sample size? With 38 predictors, your model will almost certainly be overparameterized and/or overfit. You also invite an enormous risk of collinearity among predictors.
(1) Go with regression analysis.
(2) Check you sample size, with N = 360 or more you should be fine according to
Article How Many Subjects Does It Take to Do a Regression Analysis?
(3) Check for collienarity, which is quite likely given your 38 predictors... If collienarity is given, aggregate your predictors. Factor analysis will help you with this.
(4) For estimating the relevance of your predictors try using a simple approach such as
Article A Heuristic Method for Estimating the Relative Weight of Pre...
SPSS syntax for this you will find at:
Article Two SPSS programs for interpreting multiple regression results
Happy regression
To predict values of a dependent variable (DV) from several independent variables (IVs), you could go for OLS regression, ordinal regression, or logistic regression, contingent on the type of measurement level of your DV. However, since you have many IVs, you might face a bigger problem, i.e., multicollinearity, which impacts both predictor contribution and regression coefficients’ reliability. Therefore, you should maintain only the most relevant predictors in your regression model by using approaches such as dominance analysis (Budescu, 1993) or iterative predictor weighting (IPW) PLS (Forina et al., 1999). Below are the full citations.
Budescu, D. V. (1993). Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin, 114, 542–551. https://doi.org/10.1037/0033-2909.114.3.542
Forina, M., Casolino, C., & Pizarro Millan, C. (1999). Iterative predictor weighting (IPW) PLS: A technique for the elimination of useless predictors in regression problems. Journal of Chemometrics, 13(2), 165–184. https://doi.org/10.1002/(SICI)1099-128X(199903/04)13:2%3C165::AID-CEM535%3E3.0.CO;2-Y
Good luck,
- Badges
- Science topic