You start
with a research question that includes a data collection component to the
question
   

Level of Data? 
 
 
 

Type of Data? 
 
nonparametric

 

Options to describe your data? 
Graph
Measures
Measures of Variability
Looking at relationships(Predictions)
Inferential Statistics
Measuring Differences

bar   
bar histogram  


/  bar \ histogram 

mode  median  mean  
mode  median  
mode  
Range  Range 
Range 

SemiInterquartile Range  SemiInterquartile Range  
Standard Deviation  
ZScores  
Contingency coefficient 
Spearman Rank 
Pearson's  
PhiCoefficient  Kendall Rank (small data)


Other correlation techniques where variables are:
Artificial dichotomy and Continuous = Biserial Correlation Widespread artificial dichotomy and Continuous =
Widespread Biserial
True dichotomy and Continuous =
Point Biserial
2 or more categories and 2 or more categories =
Contingency


Linear Regression  
Chisquare (Goodnessoffit test) 
2 groups and 2 conditions/independent samples
(MannWhitney U test) 
one group and infinite # of conditions:
(one sample dependent Ttest) 

Chisquare (test of independence)  2 groups and 2 conditions/dependent samples
(Wilcoxon signed ranks T test) 
2 groups and 2 conditions
( dependent Ttest ) 

2 groups and 3 or more conditions
(KruskalWallis Test) 
2 groups and 2 conditions
( independent Ttest) 

3 or more groups and 2 conditions
(KurskalWallis Test) 
2 groups and 3 or more conditions
( one way ANOVA) 

2 or more groups and 3 or more conditions: where
each group experiences each condition
(Friedman's Test 
3 or more groups and 2 conditions
(two way ANOVA) 

2 or more groups and 3 or more conditions: where
each group experiences each conditon
(Repeated Measures Anova) 
