Inferential Statistical Tests Flashcards
These types of inferential tests are used to analyze nominal and ordinal data.
Nonparametric tests
These types of inferential tests are used to analyze interval and ratio data.
Parametric tests
Parametric tests are only used when certain assumptions are met. They are…?
- Data are normally distributed
- Homogeneity of variances (i.e. variances for different groups are similar)
Non-parametric tests are sometimes choses when these are violated or the group sizes are small and unequal.
This inferential test is chosen when the data to be analyzed are all nominal.
Chi-squared
Two types:
- Single-sample chi-squared (AKA chi-squared goodness of fit test)
- Multiple-sample chi-squared (AKA chi-squared for contingency tables)
This inferential test is used when one is analyzing data from descriptive studies with only one variable.
e.g., whether or not Democrats tend to favor reducing, eliminating, or ignoring student debt (variable = choice; with three levels)
Single-sample chi-squared (AKA chi-squared goodness of fit test)
This inferential test is used to analyze data from a descriptive study that (a) has two or more variables that can’t be identified as independent or dependent variables or (b) an experimental study that has independent and dependent variables.
Multiple-sample chi-squared (AKA chi-squared for contingency tables)
This inferential test is used when a study includes one independent variable that has two levels and one dependent variable that’s measured on an interval or ratio scale.
e.g., comparing the mean mock EPPP exam scores (DV) obtained by psychologists who participated in either a live exam review workshop or an online exam review workshop (IV, 2 levels).
Student’s t-test - In this situation, the t-test will be used to compare two means.
This type of t-test is used when an obtained sample mean is compared to a known population mean.
T-test for single sample
This type of t-test is used to compare the means obtained by two groups when subjects are randomly assigned.
T-test for unrelated samples
This type of t-test is used to compare the means obtained by two groups when subjects are
- naturally related (e.g., twin studies) and assigned to different groups
- matched in pairs based on pre-test scores or and extraneous variable
- single-group, pretest-posttest design is used and subjects are paired with themselves.
t-test for related samples
This inferential test is used when:
- a study includes one independent variable which has more than two levels
- one dependent variable that’s measured on an interval or ratio scale
- the groups are unrelated
e.g., comparing the effects of CBT, ACT, and DBT (IV , 3 levels) on severity of depressive Sxs (DV) when subjects are randomly assigned
One-way ANOVA
Though one-way ANOVAs can be used when there’s a single, 2-level IV, traditionally a t-test is used for those.
We want to use the least number of tests in data analysis, because each additional test increases the probability of a Type I error.. This is called _____.
Experimentwise error rate
e.g., two tests at an alpha of .05 increases the probability of a Type I error to 10%.
In an F-ratio, the numerator is the measure of variability in dependent variable scores that’s due to treatment effects plus error and is known as _____?
Mean square between (MSB)
If this is larger than the MSW, it suggests the IV has an effect on the DV.
In an F-ratio, the denominator is the measure of variability that’s due to error only.
Mean square within (MSW)
This extension of the one-way ANOVA and is used when a study includes more than one independent variable.
Factorial ANOVA (also known as [number]-way ANOVA for the number of IV)
Factorial ANOVA produce f-ratios for each main effect and each interaction effect.
This inferential test is used when the data were obtained from a study that used a mixed design – i.e., when the study included at least one between-subjects independent variable and at least one within-subjects independent variable.
Mixed ANOVA (AKA split-plot ANOVA)
This inferential test is used to control the effects of an extraneous variable on a dependent variable by including it as an independent variable and determining its main and interaction effects on the dependent variable. When using this, the extraneous variable is referred as the “blocking variable.”
Randomized block ANOVA
This parametric test is used to control the effects of an extraneous variable on a dependent variable but does so by statistically removing its effects from the dependent variable. When using this, the extraneous variable is the “covariate.”
Analysis of covariance (ANCOVA)
This inferential test is the appropriate statistical test when a study includes:
- one or more independent variables
- two or more dependent variables that are each measured on an interval or ratio scale.
Multivariate analysis of variance (MANOVA)
This inferential test is used when a study includes one or more quantitative independent variables and the researcher wants to determine if there’s a significant linear or nonlinear (quartic, cubic, or quadratic) relationship between the independent and dependent variables.
Trend analysis
In inferential statistics, these comparisons are designated before the data is collected and are based on theory, previous research, or the researcher’s hypotheses.
Planned comparisons (AKA planned contrasts and a priori tests)
In inferential statistics, these comparisons are done when an ANOVA produces a significant F ratio.
Post hoc tests
Because of experimentwise error rates, it’s a good idea to reduce the alpha on the initial test.
This inferential measure of looks at the difference between groups in terms of standard deviation and is a way of indicating practical significance.
Cohen’s d
Cohen’s f is used when there are more than two groups being compared.
