Popular in Quantitative Methods Psych I
Popular in Psychlogy
This 3 page Class Notes was uploaded by Amy Turk on Wednesday April 27, 2016. The Class Notes belongs to Psyc-21621 at Kent State University taught by Dr. Gordon in Spring 2016. Since its upload, it has received 5 views. For similar materials see Quantitative Methods Psych I in Psychlogy at Kent State University.
Reviews for T-Statistic
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 04/27/16
PSYCH STATS POWERPOINT 9 ● t statistic ● when we set a = .05, this result will only occur at random 5% of the time ○ we are 95% sure our result didn’t occur by chance ● df = (n-1) ● standard error indicates the variability in M ● we use z-scores to test our hypothesis ● with t-statistics, we don’t have certain information ● confidence intervals = give us an idea of how well our numbers are at estimating the values ● we rarely know the SD ○ we use “s” to estimate the standard deviation ■ the sample standard deviation has sampling variability ● t-statistic = used to test hypotheses about an unknown population mean when the value of SD is unknown ○ when we do hypothesis testing we hypothesize a value of mew ● because we will be working with samples, we can substitute the values ● SD = square root of the variance ● we’re going to focus on formulas using variance ● degrees of freedom = describes the number of scores in a sample that are independent and free to vary ○ the greater the df, the better the sample variance represents the population variance ■ the better the t-statistic approximates the z- score ● the t-distribution will approximate a normal curve ○ how well it approximates depends on the df ● the greater the sample size the better ● the t-distribution has more variability than the z-distribution, especially when n is small ○ more flat and spread out William Sealy Gosset ● born in June 13, 1876 ● attended winchester college, and new college, Oxford ● graduated in 1899 and took employment at a brewery in Dublin ● Gosset was interested in the selection of the best varieties of barley ● he worked with Karl Peterson (a famous statistician) who did not find the problem interesting ● Gossett invented the t-distribution by working with random numbers ○ by far the most important advancement in statistical science that has ever occurred ● published his paper under a pseudonym ○ “a student of statistic” ■ “student’s t” ● to evaluate the t statistic from a hypothesis test, you must select an alpha level, find the value of df for the t statistic, and consult the t distribution table ● if the obtained t statistic is larger than the critical value from the table, you can reject the null hypothesis ● the critical value is the “cut-off” point at which your sample is statistically significant ○ using the t-table to find your critical value Step 1 ● determine the alpha level ● if alpha = .05, we know we will be looking for a proportion in the tain of .05 ● if alpha = .01, we know we will be looking for a proportion in the tail of .01 Step 2 ● determine the direction of the hypothesis ● if the hypothesis is one-sided, use the first row at the top of the table ● if the hypothesis is two sided, use the second row at the top of the table Step 3 ● determine the degrees of freedom and find it on the left side of the table ● df = n-1 Step 4 ● based on the df and the alpha level, find the corresponding critical value ● this is the value we need to exceed to have a significant finding ● one-sided = t will need to exceed this value in that direction (either positive or negative) ● two-sided = t will need to exceed the value in either direction Hypothesis testing with T ● the population must be normal ● the values in our sample must be independent ○ no relationship between observations ○ the same person cannot be included twice ● whenever a sample is obtained from a population, you expect to find error between M and mew ○ sampling error ● the goal of a hypothesis test is to evaluate the significance of this observed error ● the hypothesis test attempts to decide between the following ○ is it reasonable that the discrepancy between M and mew is simply due to sampling error and not the treatment effect? ○ is the discrepancy between M and mew more than would be expected from sampling error alone? ■ is the sample mean significantly different from the population? ● reject or fail to reject the null hypothesis based on alpha and df ● extreme t-values (large ratio values) have lower probability = reject ● small t-values (ration values closer to zero) have higher probabilities = fail to reject ● use cut-off t-scores from table B.2 ● the larger the value, the less probable it is to occur in the population ● extreme t-values have lower probability ○ reject ● small t-values have higher probability ○ fail to reject ● the t-distribution is more dependent on the degrees of freedom Chapter 10 ● 2 different t-tests ○ between subjects ■ independent measures ■ men vs. women ■ drug vs. no drug ■ different groups or people are compared ○ within-subjects ■ repeated measures ■ same people are compared ■ ex. before and after test ■ generally when you manipulate something ● ex. compare happiness before and after moving to Virginia
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'