Table of Statistical Techniques (Pallant, 2010)

RESEARCH_QUANTITATIVE

Re: Table of Main Statistical Techniques

Fr: Adapted from Pallant, J. (2010). SPSS Survival Manual: A step by step guide to data analysis using SPSS. Berkshire, England: Open University Press.

I.             Purpose: Exploring Relationships

QUESTION (Example)	PARAMETRIC STATISTIC	NON-PARAMETRIC ALTERNATIVE	INDEPENDENT VARIABLE	DEPENDENT VARIABLE	ESSENTIAL FEATURES
What is the relationship between gender & dropout rates from therapy?	None	CHI-SQURE	1 categorical (e.g., sex = M/F)	1 categorical (e.g., dropout / complete = Y / N)	The number of cases in each category is considered, not scores.
Is there a relationship between age & optimism scores?	PEARSON PRODUCT-MOMENT CORRELATION (r)	SPEARMAN’S RANK ORDER CORRELATION (rho/ῤ)	2 continuous (e.g., age, optimism scores)		1 sample with scores on 2 different measures, or same measure at Time 1 & Time 2
After controlling for the effects of socially desirable responding bias, is there still a relationship between optimism & life satisfaction?	PARTIAL CORRELATION	None	2 continuous & 1 continuous you wish to control for (e.g., optimism, life satisfaction, scores on a desirability scale)		(same as above)

QUESTION (Example)	PARAMETRIC STATISTIC	NON-PARAMETRIC ALTERNATIVE	INDEPENDENT VARIABLE	DEPENDENT VARIABLE	ESSENTIAL FEATURES
How much of the variance in the life satisfaction scores can be explained by self-esteem, perceived control & optimism? Which of these variables is the best predictor?	MULTIPLE REGRESSION	None	Set of ≥ continuous (e.g., self-esteem, perceived control, optimism)	1 continuous (e.g., life satisfaction)	1 sample with scores on all measures
What is the underlying structure of the items that make up the Positive and Negative Affect Scale – how many factors are involved?	FACTOR ANALYSIS	None	Set of related continuous variables (e.g., items of the positive & negative affect scale)		1 sample, multiple measures

II.           Purpose: Comparing Groups

General Research Question: Is there a statistically significant difference among a number of GROUPS?

QUESTION (Example)	PARAMETRIC STATISTIC	NON-PARAMETRIC ALTERNATIVE	INDEPENDENT VARIABLE	DEPENDENT VARIABLE	ESSENTIAL FEATURES
Are males more likely to drop out of therapy than females?	None	CHI-SQUARE	1 categorical (e.g., sex)	1 categorical (e.g., dropout / complete therapy)	You are interested in the number of people in each category, not scores on a scale.
Are males more optimistic than females?	INDEPENDENT SAMPLES T-TEST	MANN-WHITNEY U TEST	1 categorical (2 levels) [e.g., sex]	1 continuous (e.g., optimism scores)	2 groups: different people in each group

QUESTION (Example)	PARAMETRIC STATISTIC	NON-PARAMETRIC ALTERNATIVE	INDEPENDENT VARIABLE	DEPENDENT VARIABLE	ESSENTIAL FEATURES
Is there a change in participants’ anxiety scores from Time 1 & Time 2?	PAIRED SAMPLES T-TEST	WILCOXON SIGNED-RANK TEST	1 categorical (2 levels) [e.g., Time 1 / Time 2)	1 continuous (e.g., optimism scores)	Same people on 2 different occasions
Is there a difference in optimism scores for people who are under 35 yrs, 35-39 yrs, & 50+ yrs?	ONE-WAY BETWEEN GROUPS ANOVA	KRUSKAL WALLIS	1 categorical (≥ 3  levels) [e.g., age group]	1 continuous (e.g., optimism scores)	≥ 3 groups: different people in each group
Is there a change in participants’ anxiety scores from Time 1, Time 2, & Time 3?	ONE-WAY REPEATED MEASURES ANOVA	FRIEDMAN TEST	1 categorical (≥ 3 levels) [e.g., Time 1 / Time 2 / Time 3]	1 continuous (e.g., anxiety scores)	≥ 3 groups: same people on 2 different occasions
Is there a difference in optimism scores for males & females, who are under 35 yrs, 36-39 yrs, & 50+ yrs?	TWO-WAY BETWEEN GROUPS ANOVA	None	2 categorical (≥ 2 levels) [e.g., age group, sex]	1 continuous (e.g., Fear of Statistics test scores)	≥ 2 groups with different people in each group, each measure on ≥ 2 occasion

QUESTION (Example)	PARAMETRIC STATISTIC	NON-PARAMETRIC ALTERNATIVE	INDEPENDENT VARIABLE	DEPENDENT VARIABLE	ESSENTIAL FEATURES
Which intervention (math skills / confidence building) is more effective in reducing participants’ fear of statistics, measured across 3 time periods?	MIXED BETWEEN-WITHIN ANOVA	None	1 between-groups variable (≥ 2 levels), 1 within-groups variable (≥ 2 levels) [e.g., type of intervention, Time]	1 continuous (e.g., Fear of Statistics test scores)	≥ 2 groups with different people in each group, each measured on ≥ 2 occasions
Is there a difference between males & females, across 3 different age groups in terms of their scores on a variety of adjustment measures (anxiety, depression, & perceived stress)?	MULTIVARIATE ANOVA (MANOVA)	None	≥ 1 categorical (≥ 2 levels) [e.g., age group, sex]	≥ 2 related continuous (e.g., anxiety, depression & perceived stress scores)
Is there a significant difference in the Fear of Stats test scores for participants in the Maths skills groups & the confidence building group, while controlling for their scores on this test at Time 1?	ANALYSIS OF COVARIANCE (ANCOVA)	None	≥ 1 categorical independent (≥ 2 levels), 1 continuous covariate (e.g., type of intervention, Fear of Stats test scores at Time 1)	1 continuous (e.g., Fear of Stats test scores at Time 2)