Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
31 Cards in this Set
- Front
- Back
paired t-test
|
technique for determining if differences exist between two dependent-based means
|
|
What does the one sample z-test and the one sample t-test determine?
|
if our sample mean is significantly different from the population mean
|
|
What type of test would be used when pairs of participants are matched on one or more characteristics and randomly placed in two different groups (e.g. experimental and control)
|
paired t-test
|
|
What type of test would be used when twin studies (or siblings) where one twin is in Group 1 and the other is in Group 2
|
paired t-test
|
|
What is the null hypothesis for the paired t-test?
|
u1=u2 There is no significant difference in performance within the two conditions
|
|
What does the r in the paired t-test formula mean?
|
the correlation between the scores from the 2 dependent samples.
|
|
What should the magnitude of the r be for the scores to be considered dependent (i.e. related)
|
.4 or greater
|
|
What is the criteria for rejecting the null hypothesis?
|
p value is less than or equal to an alpha of .05
|
|
When is the z-test used?
|
when the sample data has a perfect normal distribution
|
|
What does the one sample z-test and the one sample t-test determine?
|
if the sample mean is significantly different from the population mean
|
|
When could the z-test be used?
|
when all sample data had perfect normal distributions
|
|
t distribution
|
series of approximations of the normal curve for different sample sizes
|
|
independent t-test, two-sample t-test, Student's t-test
|
comparing 2 sample means rather than a sample mean to a population mean
|
|
What makes up the independent variable in a independent t-test?
|
there is 1 IV with 2 levels: the 2 different groups that constitute you samples
|
|
What are the assumptions of the independent t-test.
|
randomness, normality, interval/ratio data, 2 independent groups, 10 or more participants per group, homogeneity of variance
|
|
What independent t-test assumption do you not have to worry about if the group sizes are equal?
|
homogeneity of variance
|
|
What do you do if the group sizes are not equal in an independent t-test?
|
test for the assumption of equal variance
|
|
Levene's test
|
used for testing for equal variance
|
|
what do you do if the variances between the gruops are equal in an independent t-test?
|
You can use a different formula for t that uses a "pooled" or average variance for the groups. This usually results in a lower error term and thus greater power.
|
|
When is the effect size used?
|
before a study to estimate sample size- based on an average effect size from the literature
|
|
What does effect sizehelp estimage?
|
the meaningfulness of your treatment
|
|
What can you use to get a better picture of the magnitude of the difference between two means after the study?
|
effect size
|
|
How is the effect size most often used?
|
to describe the difference between the mean of the experimental group and the control group
|
|
omega squared
|
another way of evaluating effectiveness of treatment
|
|
When should omega squared be calculated?
|
after a t-test if a significant difference is found
|
|
What does omega squared indicate?
|
the proportion of the variance in the dependent variable that is explained by the independent variable
|
|
normal distribution
|
scores on the dependent variable can be expressed as a standard score as a basis of comparison
|
|
What is the most common standard score used with a normal distribution?
|
z-score
|
|
z-scores
|
the standardized score used to produce a normal curve with mean=0 and s=1. It essentially represents a raw score expressed in standard deviation units. Can convert raw scores from a distribution into z-scores.
|
|
What must be known before raw scores can be reported as z-scores?
|
mean and standard deviation
|
|
The equations for calculating what is used by most stats programs based on z scores?
|
skewness
|