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;
23 Cards in this Set
- Front
- Back
Inferential statistics (2 points) |
Is often the next step after you have collected and summarized the data Inferential statistics is used to make inferences from a smaller group of data to a possible larger one |
|
Extrapolation (explain) |
selecting samples from the population and extrapolating what you found in the sample back to the larger population of interest |
|
Population (definition) |
a complete collection of all elements (scores, people etc.) that are to be studied |
|
Sample |
a subcollection (or subset) of elements drawn from a population |
|
Representative sample (3 points) |
The sampling distribution between the sample and population of interest must be the same The sample mean and the population mean must be similar Measures of central tendency and dispersion must be similar between sample and population of interest |
|
Making generalizations |
Infer results from the sample to the larger population Generalize some parameters for a population based on what is learned from the sample (s) from that population of interest |
|
Bivariate Analysis (3 points) |
Describes the relationship or association between two variables Test the relationship between an dependent and independent variable How the independent variable influences the outcome or dependent variable |
|
2 kinds of inferential testing |
Nonparametric test – violate some assumptions but are just as valuableOften don’t have normal distribution of scores or samples are just too small Parametric test – fulfill a number of important assumptions about inferential testPrimarily the sample is large enough or robust to be representative of the population |
|
How to select an inferential test (2 points) |
For discrete data: Nonparametric statistical testChi Squared For continuous data: Parametric statistical testT-test (compare means of 2 independent groups) |
|
Bivariate tables |
Bivariate table displays the results of the test of association between two variables Displays the results of the “joint frequency distribution between the two variables |
|
Chi Square (3 points) |
Test for association between two variables that have been organized into a bivariate table - also known as a test of independence Appropriate for nominally measured dependent variables Test of significance |
|
Confidence Level (2 points) |
Researchers must be confident that: - The results of their statistical test is due to some difference between the variables -Not due to chance or some other competing reason Researchers must choose a confidence level that will reduce the likelihood that the relationship between the variables is not due to the test itself or to some other competing reason |
|
Selecting Confidence Level (4 levels) |
95% confidence level – social sciences 99% confidence level – clinical trials 99.99 % confidence level 100% confidence level |
|
Test of statistical significance (3 points) |
Probability of Error -Always involves a chance of error -Is the degree of risk you are we willing to live with? - Risk associated with not being 100% confident that what you observed in an experiment is due to the treatment or an association exist between two variables |
|
P value - used for what? 3 levels |
Test of statistical significance (.05) (5% error) (95% confidence level) (.01) (1% error) (99 % confidence level) (.001) ( less than 1% error) (99.99% confidence level) |
|
How to make conclusions based on P values (2 steps) |
Compare the chosen confidence level with the P value calculated from the statistical test Based on this information, you either accept or reject the null hypothesis |
|
What to do with P (.0.5 or less than 0.5) |
If P is equal to or less than 0.05, data is statistically significant and so you reject the null hypothesis If P is less than 0.05, data is not statistically significant, so you accept the null hypothesis |
|
Null hypothesis (2 possibilities) |
Ho: The two variables are independent of each other (in other words, there is no relationship between them) H1: The two variables are not independent of each other (in other words, there is some influence of the independent variable on the dependent variable) |
|
Limitations of chi square (2) |
Data are at the nominal level Data are categorized into cells, and there must be data in each cell (at least 5 counts) |
|
T test (3 points relating to function) |
Test of Independent means ( two groups) T-Test finds out if there is a difference on the average scores of one (or more) variable(s) between two groups that are independent of each other By independent we mean the two groups are different from each other |
|
T test criteria (4) |
-Dependent variable (score) is a continuous variable (i.e.., midterm test score) -Independent variable is a discrete variable (eg. sex, age cohort) -Minimum sample size (N=30) -Equal variability between the groups – homogeneity assumption 95% confidence level |
|
Levine's test for equality of variance (4 points) |
Ho assumes two variances are equal. Since the significance level is greater (>) than 0.05, we do not reject Ho, meaning the variances of the two samples are equal/similar. On the right hand side of the Independent Samples Test table, it shows both results for equal variances and unequal variances. Based on the Levene’s Test on the left, we only read the data on the top row (equal variances assumed). |
|
Levine's test for equality of variance (3 more points) |
If the significance level was less (<) than 0.05, we would reject Ho, meaning the variances of the two samples are not equal/different. On the right hand side of the Independent Samples Test table, it shows both results for equal variances and unequal variances. Based on the Levene’s Test on the left, we only read the data on the bottom row (equal variances not assumed). |