Unformatted text preview:

Study Guide for Final 12 10 2011 Topics Quasi experimental design How is a quasi experiment from true experiment Notation Meaning of nonequivalent Threats to internal validity Selection threat selection bias No random assignment Often much easier to implement than random assignment Nonequivalent groups design Ex O X O O O Conclusion validity power What is conclusion validity data are reasonable Threats to conclusion validity An R represents Random and an N represents Nonequivalent The degree to which conclusions we reach about relationships in our Did we conclude there is a relationship Is that conclusion correct Concluding there is no relationship when in fact there is one o Issues with the measurements o Issues with the test setting o Issues with the respondents o Low statistical power What is statistical power Concluding there is a relationship when in fact there is not o Fishing for a significant relationship Type I error Type II error and power 1 Increasing decreases and increases power Low power is a threat to conclusion validity What are alpha and beta Alpha Chance of making a type 1 error rejecting the null hypothesis when it should not be rejected Beta Chance of making a type 2 error accepting the null hypothesis when it should be rejected What is power Odds that you will correctly reject the null hypothesis conclude there is a significant relationship result treatment effect How is related to power When is increased power is also increased What happens when is lowered or raised in terms of errors and test rigor error 2 error Lower level of more rigorous test less chance of making a type 1 Higher level of less rigorous test more chance of making a type Descriptive statistics What are descriptive statistics in general Provide basic summaries of the sample and measure Distribution Frequencies Frequency of individual values or ranges of values Central tendency mean median mode Central tendency Estimates of the center of the distribution of values Mean Average Median Middle number Mode Most frequent Dispersion range standard deviation Dispersion The spread of values around the central tendency Range Highest value minus the lower value Standard Deviation SD or s How the set of scores relates to the mean Don t need to know how to calculate SD but should know what it means What percentages of scores fall within 1 2 3 SD of the mean 1 SD of the mean 68 2 SD of the mean 95 3 SD of the mean 99 z score What is a z score Why convert scores to z scores Z score o A measure of how far an individual score falls from the mean in terms of standard deviations o Also known as standard scores normal scores How is z score related to standard deviation Formula o X X bar SD o Individual score minus mean score divided by standard deviation Correlations Be able to identify from a scatterplot whether a correlation is positive negative strong weak etc What does a 1 0 correlation mean What does a 1 0 correlation mean 1 0 Completely positively correlated 1 0 Completely negatively correlated What does a 0 correlation mean No relationship Correlation vs causation more things Correlation A mutual relationship or connection between two or Causation The relationship between cause and effect causality Inferential statistics What is meant by inferential statistics Statistics used to make inferences beyond the data Used to make inferences to more general conditions General things about all tests Most inferential statistics come from family known as general linear model GLM o T test comparison of 2 means o ANOVA ANCOVA o Regression Basic steps to performing a test State the null alternative hypothesis determine which statistical test should be used determine degrees of freedom choose alpha level calculate test statistic compare to critical value decide whether to reject null hypothesis Know the following terms Degrees of freedom critical value test statistic Degrees of freedom Critical value o The leeway for variation a test statistic has o As alpha level gets smaller the critical value gets larger o As the degrees of freedom increases the critical value gets smaller Test statistic o Number that is calculated using data from your sample that is compared to the critical value to determine whether to not reject the null How to decide whether to reject or accept null If the critical value is larger than the test statistic accept null If the test statistic is larger than the critical value reject null Should be able to identify what would be the best test to use in a particular scenario Comparing means of 2 groups T test Comparing means of more than 2 groups ANOVA Comparing impact of one variable on another Regression Determining relationship between nominal variables Chi square T test When is it used from each other Examples T tests asses whether means of two groups are statistically different o Do boys or girls differ in their enjoyment of a TV program o Did the treatment group for the intervention have a better posttest than the control group o Did the class have better scores on their posttest than they Between vs within groups t test for independent samples vs t test Also can determine significance of correlation t test for did on their pretest for dependent samples correlations Difference between t tests of independent means and t tests of dependent means T test of independent means o Compares 2 means from 2 different groups Steps to performing t test for independent means 1 State null and alternate hypotheses 2 Calculate group means 3 Calculate variance of the means SD2 4 Choose alpha level 5 Determine degrees of freedom The leeway for variation a test statistic has Used to determine the critical value as alpha level gets smaller the critical value gets larger as the degrees of freedom increases the critical value gets smaller Reflection of sample size test used For t test of independent samples df n 2 t X1 X2 var1 n1 var2 n2 6 Calculate test statistic 7 Look up critical value 8 Determine whether to reject or accept H0 If test statistic critical value accept H0 If test statistic critical value reject H0 T test for dependent means o Compares 2 means from the same group usually a pretest and posttest Regression analysis When is it used To determine if there is a significant relationship between two or more variables more sophisticated correlation because can be more than one IV or DV and can make somewhat of a causal statement What is the type of test used F test Similar procedure to performing t test o


View Full Document

UMD EDHD 306 - Study Guide for Final

Download Study Guide for Final
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Study Guide for Final and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Study Guide for Final and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?