SPSS Program Steps of Variables Data Entry and analyse

SPSS Program Steps of Variables Data Entry and analyse

STEP-1: Variables Identification and Entries in Variable View: 

STEP-2: Define Labels, Write Appropriate values for various Levels of each Item in Variable View and Select correct Measures 

STEP-3: Continue defining and copying the Values against each Variables’ Item as per Available data

 STEP-4: Look at the Data View of Defined Variables and Entries of Selected Data

STEP-5: Checking for Missing data and Entries

 STEP-6: Click on Transform Tab and Select Replace Missing Values (Drag variables to right sides of screen and press OK) 

STEP-7: SPSS will analyse all the data and fill the missing values with the most appropriate means of each variable entries 

STEP-8: Look at the Data View and new entries against each variables with missing data

 STEP-9: Adjust Decimals to Zero for all new entries in Variable View, appeared after Missing Values Analysis

 STEP-10: Complete the Decimals adjustment for Variables Items 

STEP-11: Perform Frequency Analysis for each Variables Items by Selecting Descriptive Statistics under Analyse Tab and Click on Frequencies

STEP-12: Perform Frequency Analysis for each Variables Items (Starting from Nominal Measures Items) by dragging into Variable view and Select Statistics

 STEP-13: Select Means of Variables Items and drag to Variable view for Frequencies Analyses instead of Original Entries (to avoid blank entries)

 STEP-14: Select Mean, Median, Mode for the selected Variables Items

STEP-15: Select Dispersion and Distribution analysis items from the given choices

STEP-16: Select Charts and turn on the ‘Show Normal Curve on Histogram’,Click Continue and OK to perform frequencies analysis

 STEP-17: View the results under Output Viewer Screen and make comparison between various Variables Items

STEP-18: Look at the Graphical presentation of each Variables Item to compare the histograms 

STEP-19: Perform Reliability Analysis of each Variables Items to check the validity of data (Select Scale under Analyse Tab and Click on Reliability Analysis) 

STEP-20: Select All the Measure Scales Items under Common Variable, Drag for Statistics, Select Item and Click on Continue and OK buttons for Analysis 

STEP-21: Look at the results on Output Views Screen, if Value of Reliability Alpha is ≥ 0.75, Data is valid otherwise continue to next Step 

STEP-22: Select again the Common Variables Items for Reliability Analysis, Click on Statistics and Select ‘Scale if Item Deleted’ then Press Continue & OK

 STEP-23: Look at the Analysis’ Result and watch for the highest value of ‘Alpha if Item Deleted’

 STEP-24: Go back to the Reliability Analysis Tab and delete the item with the highest value of ‘Alpha if Item Deleted’ then press OK to continue analysis

 STEP-25: Look at the Analysis’ Result. If Reliability Value ≤ 0.75 Stop here Otherwise, watch for the highest value of ‘Alpha if Item Deleted’ to Continue

 STEP-26: Continue performing Reliability Analyses for all Variables Items Set until, Alpha value ≥0.75 achieved

 STEP-27: Continue performing Reliability Analyses for all Variables Items Set until, Alpha value ≥0.75 achieved

 STEP-28: Perform Mean of Validated Variables Items (excluding deleted items to achieve Alpha value ≥0.75), Select Transform Tab, Click on Compute Variables 

STEP-29: Select ‘All’ from Function Group and Double Click on ‘Mean’ from Special Variables then Select Validated Variables separated by Comma, Give a Name to Target Variable. Click on OK to perform this analysis. Look at the Result Screen and confirm all Alpha Values are now greater than 0.75 

STEP-30: Perform Step-29 for all Variables (Select common variable Items and give a unique name to identify and differentiate from other variables data 

  

STEP-31: Look at the Variable View, New Variable entries are generated by SPPS for Variable data Sets. (One for each dependent and independent variables) 

STEP-32: Populate all new variable items following all previous steps and initiate validation check (Click Analyse tab, Click Correlate and then Select Bivariate)

 STEP-33: Select all Main Variables (Dependent and Independent) and Drag towards Variables Screen, Press OK for further process

STEP-34: Go to Output View for results. You will have one Correlation table showing strength of variables’ relations (-ve and +ve)

* 0.000 Sig. (2-tailed) means 0% here (very strong relation) while the result is statistically significant at the 5% (0.050) level. If P ≥0.5 then we fail to reject Null.

** If we increase Job Satisfaction by 100% then job commitment will be increased by a proportion of 0.796. Less than 0.3 value indicates a week correlation relation between any two variables.

STEP-35: Now press Analyse Tab, go to Descriptive Statistics and Select Descriptive for checking means and deviations. Select Main Variable only and press OK 

STEP-36: Look at the Output Results of Descriptive Analysis and find out the mean of overall responses against each Variable

Pearson Correlation must be greater than 0.3 to show the strength of predictors/ independent variables over dependent variable (irrespective of the –ve or +ve relation between independent and dependent variables). In the same time, this Pearson Correlation value must not be more than 0.7 for two predictors. Lesser value for two predictors/ independent variables means two different and distinct variables otherwise, two variables in fact are behaving like one variable.

  

STEP-37: Check frequencies of each Main Variables. Press Analyse Tab, go to Descriptive Statistics and Select Frequencies. Drag all Main Variables for Analysis

STEP-38: Look at Output view to investigate the details of Demographic items, Variable’s Responses Breakdown, Means, Deviation, Minimum and Maximum 

STEP-39: Press Analyse Tab, Press Regression and Select Linear Tab to check relation between various Variables. Select Dependent & Independent Variables

1-      There is +ve or –ve relationship between two variables.

2-      Correlation between two variables (independent and dependent) is Zero

3-      There is no supported correlation/ relationship between two variables i.e. unrelated variables.

STEP-40: Look at the Output Result View to either Reject or Validate Null Hypothesis. R Value higher 0.7 means higher similarity between two Variables

Delete variables with higher “R” values as responses indicate these variables similar in nature i.e. multicollinearity nature of two variables and redundancy of two predictors. Logically either to combine those two predictors to form a composite variable or eliminate one variable. “R Square” generally must be between 0.2-0.3 to validate Hypothesis. “R Square” indicates that selected independent variables has impact of 29.7% on dependent variable (significant relation b/w independent and dependent variables).

STEP-41: Now Perform T-Test between Demographic Items and Variables to see the responses (Press Analyse Tab, Compare Means and then Select Independent

STEP-42: Select Demographic Item like Gender or Age, drag to Grouping Variables, Define Groups as per its originally defined value (1, 2 or as you have defined)

 STEP-43: After selecting the Grouping Variables and defining it, Select One Main Variable as Test Variable and Press OK for further analysis

  

STEP-44: Look at the Demographic Grouping Variables such as Females and Males responses (Means, Sig) against Main Variable such as Job Turnover

Check the Questionnaire items 2 and 3 under Variable Turnover to check the responses of overall males and females for this Main Variable. As Sig. ≥ 0.05 therefore, females and males both don’t have significant relation for Job Turnover variable.