Practical Exam Flashcards
When do we use the Kolmogorov-Smirnov normality test?
for sample sizes >-50
When do we use the Shapiro-Wilk normality test?
for n<50
What does it mean if the significance (p) is less than or equal to 0,05?
reject Ho.
NOT normal distribution
What does it mean if the significance (p) is bigger than 0.05?
accept Ho.
Normal distribution :)
Which type of analysis do we use for comparing nominal values with nominal values?
Chi-Square
Which type of analysis do we use when comparing nominal to ordinal?
Chi-Square
Which type of analysis do we use when comparing Ordinal to Ordinal?
Chi-Square
Which type of analysis do we use when comparing Ordinal with Quant. Normal?
Spearman
Which type of analysis do we use when comparing Ordinal with Quant. Non normal?
Spearman
Which type of analysis do we use when comparing Quant. Normal with Quant. Normal?
Pearson
Which type of analysis do we use when comparing Quant. Normal with Quant. non normal?
Spearman
Which type of analysis do we use when comparing Quant. Non normal with Quant. Non normal?
Spearman
How do we do Chi-Square on SPSS?
Analyze - Descriptives - Crosstabs
How do we interpret Chi-Square?
the closer value of statistic is to 0, the lower the association
can we interpret the direction of the association with Chi-Square?
no
what does the null hypothesis in chi-square (x² =0) mean?
that there is no association
-> is it statistically significant (p<0.05)?
How do we do a Spearman/Pearson test?
Analyze - Correlate - Bivariate
-> choose Pearson or Spearman
How do we interpretate the pearson and spearman test?
positive correlation index (+) indicates direct relationship. (-) indicates inversive relationship
How do we interpretate the strength of the association in Pearson/Spearman test?
weak (0,1-0,3)
moderate (0,3-0,5)
strong (>0,5)
Based on what do we analyze if we accept or reject the null hypothesis in Spearman/Pearson?
based on the significance level
What does the Significance level determines?
the probability that observed correlation is due to chance
What indicates a strong and reliable relationship?
low significance level (<0,05) and high coefficient (>0,5)
How do we calculate dispersion measures (variance, standard deviation, range, standard error, etc.)?
Analyze - Descriptives - Descriptives
How do we interpretate skewness?
comparing value of the statistic with its standard error
double the std. error
when statistic less than twice the error: symmetric
greater than twice the error: asymmetric
How can we interpretate kurtosis?
if greater than 0.5: leptokurtic
if between -0.5 and 0.5: mesokurtic
if less than -0.5: platykurtic
What do we do if only looking for a selected subject (e.g.: only males, older than 25 years)?
data - select cases - if the condition is satisfied
How do we compare two groups separated by the categories of our choice in the same table?
data - split file - compare groups
How do we make the results of the analysis appear separated by the categories of a variable of our choice in different sections?
data - split file - organize output by variables