Seminar 2: Correlation Analysis using SPSS
SESSION OBJECTIVES
The focus of this session is to:
Teach how to complete a correlation in SPSS.
Explain how to write up your correlation results.
RUNNING A CORRELATION
The data collection will utilize two scales:
Honesty-Humility: A subscale derived from the HEXACO questionnaire (Lee & Ashton, 2004).
Morality: A subscale from the Temperament and Character Inventory (Cloninger et al., 1994).
Formulating an expectation regarding the relationship:
Consider whether it will be a Positive or Negative correlation.
The proposed hypothesis is:
There will be a positive relationship between morality and honesty-humility.
Definition of Hypothesis:
A hypothesis is a prediction made at the start of a quantitative study, stating expected findings usually based on previous literature.
HONESTY-HUMILITY SCALE
Questionnaire items rated on a scale of 1-5:
1 = Strongly Disagree
2 = Disagree
3 = Neutral
4 = Agree
5 = Strongly Agree
Sample questions include:
I wouldn't use flattery to get a raise or promotion at work, even if I thought it would succeed.
If I knew that I could never get caught, I would be willing to steal a million dollars.
Having a lot of money is not especially important to me.
I think that I am entitled to more respect than the average person.
If I want something from someone, I will laugh at that person's worst jokes.
I would never accept a bribe, even if it were very large.
I would get a lot of pleasure from owning expensive luxury goods.
I want people to know that I am an important person of high status.
I wouldn’t pretend to like someone just to get that person to do favours for me.
I’d be tempted to use counterfeit money, if I were sure I could get away with it.
MORALITY SCALE
Questionnaire items rated on a scale of 1-5:
1 = Strongly Disagree
2 = Disagree
3 = Neutral
4 = Agree
5 = Strongly Agree
Sample questions include:
I listen to my conscience.
I try to fool others.
I act according to my conscience.
I believe that the end justifies the means.
I like harmony in my life.
I misuse power.
I return extra change when a cashier makes a mistake.
I do the opposite of what is asked.
I stand behind my actions.
I care about justice.
SPSS OVERVIEW
SPSS (Statistical Package for the Social Sciences) can:
Analyze data using inferential statistics.
Produce a wide range of descriptive statistics, including Mean and Standard Deviation.
Importance of input data format:
SPSS is stringent in data input.
OPENING SPSS
Step-by-step instructions:
Open SPSS from the desktop.
Create a new data set.
INPUTTING DATA INTO SPSS
VARIABLE VIEW
Explanation of view modes:
Data View: Displays data.
Variable View: Allows managing data properties.
Inputs:
Naming columns (including participant IDs and response variables).
DATA PARAMETERS IN VARIABLE VIEW
Important parameters for each column:
Name: Column names to reflect variables.
Type: Specifies whether information is numeric/categorical.
Decimals: Indicates how many decimal places to display (e.g., 2 for 10.25).
Values: Assign numeric codes to written categories (SPSS requires numeric representation).
Measure: Essential for statistical analysis; options include:
Nominal - Categorical data without meaningful order.
Ordinal - Categorical data with meaningful order.
Scale - Numerical data.
SCORING THE QUESTIONNAIRES
Participants contribute responses under their respective rows.
Reverse Coding: Identifies items requiring a reversed scoring approach denoted by ‘R’ in item numbers:
Honesty-Humility: 1, 2R, 3, 4R, 5R, 6, 7R, 8R, 9, 10R.
Morality: 1, 2R, 3, 4R, 5, 6R, 7, 8R, 9, 10.
To score:
The score for each subscale is the average of standard and reversed items.
USING SPSS FOR REVERSE SCORING
Actions for creating new variables:
Select 'Recode into Different Variables' to avoid overwriting original values.
Specify the new variables as reversed versions (e.g., HH2R for question 2).
Detail on setting original and new value correspondence for reverse scoring:
Original 1 becomes new 5, 2 becomes new 4, etc.
COMPUTING AVERAGE SCORES
HONESTY-HUMILITY SCORE
Calculate honesty-humility average:
Average of questions 1, 3, 6, 9 and reversed questions 2, 4, 5, 7, 8, 10. Formula:
MORALITY SCORE
Calculate morality average:
Average of questions 1, 2R, 3, 4R, 5, 6R, 7, 8R, 9, 10. Formula:
TESTING HYPOTHESIS
Running the correlation test:
Reiterate hypothesis:
Expect a positive relationship between morality and honesty-humility.
ASSUMPTIONS FOR CORRELATION
Criteria for data suitability:
Data must be continuous.
The relationship must be linear.
No significant outliers should exist.
Data should be normally distributed.
SPSS can aid in testing these assumptions via manual provided.
PERFORMING CORRELATION IN SPSS
BIVARIATE CORRELATIONS
Paths to executing correlation:
Navigate via Analyze → Correlate → Bivariate.
Choose Pearson for parametric correlation; select one-tailed for directional hypothesis.
GENERATING OUTPUT CORRELATION
The output includes:
Descriptive statistics (means and standard deviations).
Correlation coefficients with significance levels.
Example output content:
Honesty-Humility Mean: 3.1 (SD: 0.46188)
Morality Mean: 2.93 (SD: 0.38312)
Correlation coefficient: r = -0.107, indicating a weak negative relationship.
Significance value: p = 0.385; indicating no statistical significance as it exceeds .05.
WRITING RESULTS
INTERPRETATION STEPS
Descriptive statistics interpretation:
Examine means and determine average satisfaction on scales (both hover around midpoint).
Graph inclusion:
Visual representation recommended through Scatter/Dot plots.
Follow steps to add a best-fit line and manipulate graphical elements.
Correlation interpretation:
Access correlation coefficient, significance levels, and effect size.
Final correlation phrase: reports non-significant, weak, negative correlation (e.g., r = -0.107, p = 0.385, R2 = 0.01).
CONCLUDING REMARKS
Summarized session objectives, ensuring students grasped the methodology to run correlations in SPSS and articulate the results effectively.