NHST
T Test basics
Assumptions
T Test in SPSS
Random
100
True or false: In NHST, we're trying to prove the null hypothesis is false.
What is False. We never prove anything in statistics. We are trying to reject the null hypothesis.
100
T tests are used to compare _________.
What is groups or means
100
True or false: most social science variables are probably not truly interval scales
What is True
100
True or false: Running a T test is found under the compare means tab.
What is True
100
The command we always use in SPSS after completing any analysis.
What is paste
200
The traditional cutoff for statistical significance
What is p < .05
200
Used when there are two experimental conditions and different participants assigned to each condition
What is independent samples t test
200
Parametric tests are also known as _______ _________.
What is Parametric tests are also known as inferential statistics.
200
The value that tells us if our test statistic is significant or not.
What is p value
200
Correlations range from ___ to ___.
What is -1 to 1
300
________ error: in reality, an effect exists, but it wasn't detected in our study. ________ error: We observed an effect in our data when, in reality, it doesn't exist.
What is Type II error: in reality, an effect exists, but it wasn't detected in our study. Type I error: We observed an effect in our data when, in reality, it doesn't exist.
300
In a simple T test, there is __ predictor that is manipulated in only __ ways and has __ outcome variable.
In a simple T test, there is 1 predictor that is manipulated in only 2 ways and has 1 outcome variable.
300
One participant’s scores on a dependent variable don’t effect those of another participant
What is independence
300
In order to check homogeneity of variance, we use this test.
What is Levene's test
300
categories without order (e.g. male/female, religious preference, meal preference)
What is nominal
400
Computing effect sizes for a series of studies that investigated the same research question, and taking an average of those effect sizes.
What is meta-analysis
400
Expose the same people to different experimental manipulations at different points in time
What is repeated measures/within-subjects
400
Rules of thumb: Skew >___ is bad Kurtosis >___ is bad
What is Rules of thumb: Skew >3 is bad Kurtosis > 10 is bad
400
Calculate degrees of freedom for our sample if group 1 has 20 participants, and group 2 has 18 participants.
What is 20 + 18 = 38 - 2 = 36
400
an estimate of total variability (spread) of a set of observations around a parameter (such as mean)
What is sum of squares
500
Name two problems with NHST
What is We never really prove the null (or the alternative) hypothesis. The null is never really true (e.g. never exactly zero effect). Creates unfounded devotion to .05. (All or none thinking) Statistical significance isn’t practical/clinical significance Significance levels are highly influenced by N!
500
3 assumptions for independent T tests
The two groups are independent of one another. The dependent variable is normally distributed. The two groups have approximately equal variance on the dependent variable.
500
Name all 4 assumptions of parametric tests
What is Interval or ratio data Independence Normal distribution Homogeneity of variance
500
True or false: If Levene's test p = .03, then we assume equal variances.
What is False. If Levene's test is significant, then equal variances are not assumed.
500
Correlation coefficient r = __ (small effect) r = __ (medium effect) r = __ (large effect)
What is Correlation coefficient r = .10 (small effect) r = .30 (medium effect) r = .50 (large effect)
M
e
n
u