RMS 618 Module 1 - SPSS Commands

docx

School

Trident University International *

*We aren’t endorsed by this school

Course

608

Subject

Statistics

Date

Feb 20, 2024

Type

docx

Pages

Uploaded by DoctorBraveryWalrus7988

Module 1 - SPSS Commands Correlation and Simple Linear Regression SPSS Commands Preparation for Analysis SPSS Commands Check assumption of normality/homoscedasticity (for regression) Select Analyze -> Regression -> Linear. Enter DV and IVs. -Select “Plots.” 1) Select ZRESID (or SDRESID) as the y variable and ZPRED (standardized predicted values) as the x variable. Select histogram and/or normal probability plot. Check assumption of collinearity (for regression) Select Analyze -> Regression -> Linear. Enter DV and IVs. -Select “Statistics.” Select collinearity diagnostics. Select continue. The following suggest a multicollinearity problem: condition index >15, VIF>4, Tolerance < .2 Create dummy codes for variables (for transforming categorical variables to dichotomous variables in regression) Select Transform -> Recode into different variables. Select variables to be transformed. Enter name and label. Press change and then "old and new values." Enter old value (the value to be changed) and new value (the value of the transformed variable). Press "Add" to add the recoding to the list. When all the recodings have been added, click on the "Continue" button and then the "OK" button. Type of Analysis SPSS Commands Descriptive Analysis #1 Select Analyze -> Descriptive Statistics -> Frequencies. -Display frequency tables - mark this if you are examining categorical variables -Select “Statistics” and check mean, median, standard deviation, skewness, and kurtosis (to check for normal distribution)

-Charts: select this option to examine frequencies of continuous variables Descriptive Analysis #2 (more detailed - more focus on outliers and assumption checking) Select Analyze -> Descriptive Statistics -> Explore. Enter DV and factors. -Select “Statistics” and check descriptives -Boxplots: mark this to examine if there are outliers Correlations Select Analyze -> Correlate. For bivariate correlations, select Pearson (for interval or ratio data) or Spearman (for ordinal data). Simple Linear Regression Select Analyze -> Regression -> Linear. Enter the dependent and independent variables. Then click ok. Multiple Regression Select Analyze -> Regression -> Linear. Enter DV. -Select one of the following options: 1)Forced entry: Select enter for method. Add all independent variables. Then click ok. 2) Stepwise: Select stepwise for method. Add all independent variables. Select statistics and check R squared change. Then click continue and ok. 3) Hierarchical: Select enter for method. Enter first group of independent variables; then select next. Repeat for subsequent groups of variables until all variables have been entered. Select statistics and check R squared change. Then click continue and ok. Logistic Regression Select Analyze -> Regression -> Binary logistic. Enter dependent and independent variables. -Select the categorical button. Enter the categorical variables (or ordinal variables with a few categories). Click continue. -Select the options button and check classification plots and Homer-Lemeshow goodness-of-fit test. Click continue.

-Select the logistic regression method. Then click continue and ok. MANOVA/MANCOVA Select Analyze -> General Linear Model -> Multivariate. Enter dependent variables, categorical predictor variables in fixed factors, and scale predictors in covariates. -Select options. Mark descriptive statistics, estimates of effect size, and homogeneity test. -Select Post Hoc. Move factors to the Post Hoc test box. Select Bonferroni Factor Analysis Select Analyze-->Data Reduction-->Factor. -Select Descriptives. Check univariate descriptives, initial solution and KMO and Bartlett's test. -Select Extraction. Select principal components method and check correlation matrix, unrotated factor solution, scree plot, eigenvalues over 1. -Select Rotation. Check Varimax. * Tip: Click on the help button for the statistical test you are running within SPSS. Then select show me (which will open a tutorial on how to run the test and interpret the output). SPSS Output: Interpretation of Coefficients in Linear Regression The following output was obtained for Subject A as the independent variable and Subject B as the dependent variable. The following output may be interpreted as follows: For each increase of 1 unit of score on Subject A, there is a .641 increase in the score of Subject B. Coefficients Unstandardized Coefficients Standardized Coefficients t Sig. Model B Std. Error Beta 1 (Constant) 16.237 4.078 3.982 .001

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version