Chapter 9 lectures 8,9,10 Flashcards
T-tests are used when you are?
Comparing differences between 2 groups on a DV with a scale-level of measurement.
What does an independent samples t-test compare?
It is used for comparing 2 groups that are indeed independent–>different subjects in each group —->between-group design
Used to compare the means of two groups when the DV has a scale-level of measurement
What group design is used when you are doing an independent samples t-test?
between group
What group design is used when you are doing a paired samples t-test?
Within-group
What does a paired samples t-test compare?
is used for comparing scores obtained from the same subjects at 2 different times
Used when data are paired or linked in some way, such as in experimental situations with pre-test scores and post-test scores
What are the three assumptions of the independent samples t-test?
Variances are equal
Normally distributed
Independent scores for each group (one person’s score doesn’t relate to a score in the other group)
What does Levene test assess?
assesses whether the variance of the dependent variable is equal for the two groups in the population. Also referred to as “Homogeneity of Variance”
If Levene’s test is NOT significant (Sig value > .05), then we can
Accept the null?
Reject the null?
we can NOT reject this null so fail to reject the null
If Levene’s test IS significant (Sig value <.05), then we can?
accept the null
reject the null
Reject the null
If Levene’s test IS significant (Sig value <.05), the assumption or normality ______ violated
IS Violated
If Levene’s test IS significant (Sig value <.05), Equal variance is ________
equal variances NOT assumed
If Levene’s test is NOT significant (Sig value <.05), Equal variance is ________
equal variances are assumed
What are the 2 assumptions of the paired samples t-test?
The Independent variable is dichotomous.
The DV is normally distributed
Is there a Leven’s test for paired sample t-tests?
SPSS doesn’t automatically do a Levene’s test because there are no
need to test for equal variances as the groups are related and similar.
The measure of DV Ordinal, what test do you choose for between group design?
Mann- Whitney U
The measure of DV Ordinal, what test do you choose for within group design?
Wilcoxon
The measure of DV is nominal or dichotomous, and it is difference question what test do you choose for within-group design?
Mcnemar
The measure of DV is nominal or dichotomous what test do you choose for between-group design?
Chi-Square
The Wilcoxon matched-pairs test is used to compare?
Medians (or mean ranks) of 2 repeated groups
The Wilcoxon test is used when the DV is ________
Ordinal
The McNemar test is used to compare
Used when comparing two related groups on a DV with a dichotomous level of measurement
What is the DV measurement if you are running a McNemar test?
Dichotomous
Is McNemar for between-group or within-group design?
Within
Effect direction for T-test, Mann Whitney U, Wilcoxon, McNemar, or Chi-Square we use?
We use the group means for difference questions/ Which group did beter
What is sampling distribution?
Sampling distribution is the set of different scores (or statistics) that results from repeated replications of the same study using different samples of the same size from the same population
2 Assumptions of sampling distributions are?
Normally distributed (resemble a normal curve)
The mean, median, and mode of zero. It is a theoretical distribution
What is a standard error in an independent t-test?
Standard Error is a measure of the amount of variation in a sampling distribution. It is the standard deviation of the sampling distribution.
What does a t-score tell us?
T-scores tell us how many standard deviations the difference between sample means is from the mean of the sampling distribution, and in which direction.
What is a confidence interval?
Confidence Intervals are a range of estimates within which the population parameter (e.g., mean) is likely to be found.
CI contains the population parameter.
What % do researchers use for the confidence interval?
95%
The probability of the CI capturing the population parameter is called the
_________
confidence level
The confidence interval makes us sure that 95% of the time an experiment is repeated or replicated, the true mean difference will fall between the ______ and ______
Lower and Upper confidence interval
Anytime the 95% confidence interval for the difference between 2 groups crosses zero, we will _______ the null hypothesis
NOT reject the null (fail to reject the null)
Anytime the 95% confidence interval of the difference between 2 groups does NOT cross zero, we will ______ the null hypothesis
reject the null
For independent t-tests, the effect direction is checked based on the difference in ________regardless of the T-test being significant or not
means
A one-way ANOVA is used when?
There is only 1 IV which has three or more levels and 1 scale DV
Which research design is being used in a one-way ANOVA?
Between-group
What are the assumptions of a one-way ANOVA? 3 of them
groups on the IV are unrelated
the dependent variable is normally distributed
Variances of the DV in the groups on the IV should be equal.
What does the p-value in an ANOVA tell you?
Whether or not there are overall differences between the groups on the IV
What does the p-value in an ANOVA not tell you?
which groups are different from each other
In an ANOVA, How do we find what groups were significant significant vs not?
We have to run posy hoc comparison test
When are post hocs run?
When the ANOVA is significant
What are the two types of post hocs?
Tukey and Games-Howell
When do we use the Tukey Post Hoc?
Use Tukey post hoc when the p-value for Levene’s test is not statistically significant (p > 0.05)
When do you use Games-Howell?
Use Games-Howell when the p-value of Levene’s test is statistically significant (p < 0.05)
Tukey’s HSD is used when assumptions of homogeneity of variance is__________
met
Games-Howell is used when assumptions of homogeneity of variance is__________
violated
If there is a negative number in the mean difference collum of a post hoc test, it means that the __________
first DV is is more then the second
mean plot graphs are helpful with?
visualizing group differences
Mean plot graphs can tell us to accept or reject the null.
True/False
False there is no p-value
Kruskal-Wallis H is used when the DV is _______, and the IV has _______ levels
Ordinal and the IV has three or more levels
Chi-Square is used when the DV is _______ and the IV has _______ levels
Nominal or Dichotomous and has three or more levels
General Linear Model (GLM) is a parametric test for_______ research design?
Within-group
General Linear Model (GLM) is a parametric test for_______ research design
Within-group
When is the General Linear Model used?
Used when DV is scale and IV has three or more related groups
The Friedman (Two-Way) ANOVA Test is a nonparametric test for________research design
Within
The Friedman (Two-Way) ANOVA Test is a nonparametric test for________research design.
Within
The Friedman (Two-Way) ANOVA Test is used to compare?
The differences between three or more related groups on the IV
And the DV is ordinal
COCHRAN Q Test is a non-parametric test for ________ research design
within-group
COCHRAN Q Test is used when the DV is ______ and the IV has ____ or more levels or groups
Nominal and the IV has 3 groups
What are the two assumptions of the COCHRAN Q Test?
The sample is randomly selected from a population
Responses (or scores) of subjects are categorical or dichotomous
When a research design is within group the IV is almost always _____
time
What is another name for a two-way ANOVA for between-group
factor
When is a two-way anova used? 3 of them
Used when there are 2 IVs and 1 DV
It is a different question
Between-group design
When you have a two-way ANOVA, how many hypotheses do you have?
3
What is the 3 hypothesis in a two-way ANOVA?
- No main effect of the first IV
- No main effect of the second IV
- There is no interaction between the two IV’s
What is an interaction between variables?
When 2 variables cross each other
Post Hoc will only work with an Anova if there are more than _____ groups or levels of the IV
3
Look at the _____ differences to look at the differences between groups in an ANOVA
Mean
Partial ETA squared is also used for _______ in a two-way ANOVA
There are two of them
Effect size
&
Adjusted R square, so it shows the percentage of how much the DV can be explained by the IV
When are a one-way repeated measures ANOVA used?
When it is within-group design
&
When 1 IV has three or more related groups and 1 DV, which is scale
When are Two-way Repeated Measures ANOVA used?
Within-group design
&
When 2 IVs have 2 or more related groups and only 1 DV that is scale.
When is Two-way ANOVA for Mixed Designs used?
When there is a combination of within-group and between-group design
The # of ways an ANOVA is is based on the number of _______
IVs (Three-way ANOVAs, Four-way ANOVAs, Five-way ANOVAs, etc.)
When is a MANOVA (Multiple ANOVA) used?
For designs with 2 or more DVs considered simultaneously – the DVs get mathematically combined into 1 DV for analysis purposes. There can be 1 or more IVs as factors.
When is an ANCOVA used?
Between-group ANOVAs (one-way, two-way, etc.), statistically controlling for confounding variables (covariates)
Can 2 IVs have interaction and cross each other and not be significant?
Yes, this is due to chance
There will be no interaction if the group means_________
don’t intersect
What is another name for Adjusted R squared
Percentage of Error. Tells us how much the DV can be explained by the IV