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;
13 Cards in this Set
- Front
- Back
Rules of thumb
|
- Always report total # of cases
- Do not calculate proportion and percentage for <50 cases |
|
Frequency distributions
|
- Tally of raw number of responses
- One-to-one match between response categories in the surey and in the table - Ordinal categories treated as independent units |
|
Proportion
|
- Frequency of repsonses relatitve to the total
- Proportion for total : 1 |
|
Frequency and percentage
|
- Mainly for nominal and ordinal (categorical) variables
- Always report percentage before frequency: 75% (55) - Not commonly used with interval variables - NOT used for ratio variables |
|
The median assumes an _____ distribution of the data above and below midpoint and is NOT used for _____ data
|
equal; nominal
|
|
Mean
|
- Usually applied to continuous data
- Reported with standard deviation - Not appropriate for nominal, ordinal variables |
|
When is median more appropriate than the mean?
|
-When dealing with outliers in the data set
- When open-ended categories appear in a data grouping ("over $100,000") |
|
Data thas has been skewed right...
|
...has the smallest number with the highest frequency; mode < median < mean
|
|
Data that has been skewed left...
|
...has the largest number with the highest frequency;
mean < median < mode |
|
Cross-tabulation is a relationship between _______
|
two categorical variables
|
|
If sig (2-taliled) is less than ___, we call it significant because that number represents ________
|
.05; the likelihood of the mean differnce being attributed to pure chance or error
|
|
Correlation is the relationship between ____
|
two continuous variables
|
|
T-tests express the relationship between ______
|
one continuous and one dichotomous variable
|