PSY 2310: Module 13
Which alpha level has a lower risk of making a Type I error?
.01
Which alpha level has a lower risk of making a Type II error?
.05
Use the "Independent Samples Test" output to find the Sig. value for Levene's test and then determine if the variances are similar or NOT similar. The Sig. (p) value for Levene's test is ________ and therefore the variances for the verbal and visual groups __________.
.676; are similar (i.e., equal)
If the null hypothesis is true, the z for a sample mean, single-sample t test, related samples t test, and the independent samples t test all expect an obtained value close to __________.
0
When computing the numerator of the independent samples t test, the population mean difference (i.e., μ1-μ2) will always be __________.
0
The p value for this t test was __________.
0.097
Compute the effect size for this study.
1.38 mean 1 - mean 2 divided by the square root of the pooled variance value
Use the "Group Statistics" output to determine the mean and standard deviation of the verbal learner group. The mean and standard deviation of the verbal learner group were __________.
20.1667 and 1.32916, respectively
For the above study, compute the pooled variance, SDp2.
3.17
If the Sig. for Levene's test was less than .05, the correct obtained t value would have been __________.
9.849
When entering data for an independent measures t test, you should have __________.
A column for the IV and a column for the DV.
Assignment 12
Assignment 12
Chapter 10 Reading Questions
Chapter 10 Reading Questions
Based on the results for Levene's Test of Equality of Variance, will you use the t test values for "Equal variances assumed" or the t test values for "Equal variances not assumed"?
Equal variances assumed because the value for f is greater than 0.05 then use equal variance assumed if it were lower then we use equal variances not assumed
A researcher completed an experiment to determine if getting a good night of sleep helps college students taking math classes by obtaining a sample of N = 30 students enrolled in a math class at a local community college. One group was composed of 15 students who regularly slept six or more hours every night and a separate sample of 15 students who regularly slept less than six hours every night. Each student's scores on a variety of tests were combined into a single measure of math performance with higher scores indicating better performance. Test for differences in math test scores based on having a more or less than six hour of sleep every night, using a two-tailed test with an alpha level = .05. What type of analysis should be used to test this data?
Independent Samples t Test
Which significance test should you use to determine if the differences between two unrelated samples is likely to be due to sampling error?
Independent samples t test
Use the "Group Statistics" output to determine the M and SD math scores for the group of college students who regularly sleep less than six hours.
M = 75.83, SD = 5.18
Use the "Group Statistics" output to determine the M and SD math scores for the group of college students who regularly sleep six or more hours.
M = 82.50, SD = 4.47
QUIZ
QUIZ
What decision will you make regarding the null hypothesis (H0)?
Reject H0 because the t value falls outside of our sig value for two tailed t = 3.893 sig for two tailed = .001
Use the "Independent Samples Test" output to find the Sig. value. for Levene's Test for Equality of Variance, and use that to determine if the assumption for homogeneity of variance was violated.
Sig. = .519; Have NOT violated the assumption for homogeneity of variance. we don't use the p-value (p-value is the value under the sig (two-tailed)) in this case, we use .519 A sig-value less than 0.05 indicates a violation of the assumption.
Identify which of the following statements best represents a description/interpretation of the results, including effect size, in APA format.
Students enrolled in a math class at community college, who regularly slept six or more hours a night, were found to have significantly higher average math performance scores (M = 82.50, SD = 4.47) versus students who regularly slept less than six hours a night (M = 75.83, SD = 5.18), t(28) = 3.89, p = .001, d = 1.38. Students who regularly sleep six or more hours a night, on average, have math performance scores 1.38 standard deviations above students who regulalry sleep less than six hours per night.
What is the verbal research hypothesis (H1) for this study?
The mean math performance score in the population of community college students who regularly get less than 6 hours of sleep a night is different from the mean math performance scores in the population of students who regularly get more than 6 hours of sleep a night. opposite of the verbal null hypothesis
What is the verbal null hypothesis (H0) for this study?
The mean math performance score in the population of community college students who regularly get less than 6 hours of sleep a night is not different from the mean math performance scores in the population of students who regularly get more than 6 hours of sleep a night. it is not different because U1 = U2
Which of the following is the best summary of the two-tailed research hypothesis?
The people in the verbal description group will have different memory scores than people in the no verbal description group.
Which of the following values can NEVER be negative?
The standard error of the mean difference.
The independent samples t test can be thought of as the ratio of the observed difference between two sample means divided by an estimate of sampling error.
True
What type of hypothesis testing error might you have made and why? Hint: If you Reject H0 you can only make a Type I Error, and if you Fail to Reject H0 you can only make a Type II Error.
Type I error; may have rejected a true H0
Researchers need to be very cautious when interpreting null results because __________.
a null result might occur because of a problem with the study's experimental procedure
The estimated standard error of the mean difference is __________.
an estimate of how different the two sample means are expected to be due to sampling error
When the obtained t value is NOT further from zero than the critical value an the null hypothesis should __________.
ans = not to be rejected
An independent samples t test can be used to compare differences between that __________.
both of the above a. are created by the researcher by providing different IV levels b. already exist in different populations in people
For the above study, identify the df, critical region when testing using α = .05 in two tails.
df = 68, t</ -1.9955 and t>/ 1.9955
To get a one-tailed p value, you must __________.
divide the two-tailed p value in half
Calculate the pooled variance value for this study.
https://www.socscistatistics.com/descriptive/variance/default2.aspx estimate the ans from this web to the one closest to options available 23.37
The Levene's test will help you to determine __________.
if the two conditions have similar variances or variances that are very different (i.e., if the homogeneity of variance assumption is met or violated)
When determining if an effect size is small, medium, or large, you should __________.
ignore the sign of the computed effect size (i.e., use its absolute value)
The numerator of the independent samples t test is the difference between __________.
not a sample mean and a population mean try two sample means
When writing your results, if you failed to reject the null hypothesis and yet the effect size was medium or large, you should __________.
point this out to the reader so that your report does not mislead them
When you are doing a one-tailed t test, the critical region is always on the side of the t distribution that is predicted by the __________.
research hypothesis
For the above study, compute the standard error of the mean, SEM.
root of SD1 square/ N1 + SD2 square/N2
The independent samples t test is a ratio of the difference between two sample means over an estimate of __________.
sampling error
Whenever you failed to reject the null hypothesis and yet the effect size is medium or large, you should conclude ___________.
that your sample size was too small and you should rerun the study with a larger sample size
You use an independent samples t statistic when __________.
the IV defines two independent samples, and the DV is measured on an interval/ratio scale
When computing the standard error of the mean difference, the equation calls for using __________.
the sample sizes from each group
A therapist would like to know if therapy for anxiety can be treated as effectively through teleconferencing as it can through traditional face-to-face therapy or whether one type of therapy would be more effective. After 3 months of therapy, samples of 35 clients from each of the two groups completed the Generalized Anxiety Disorder 7-Item Scale. Responses ranged from 0 to 21, with higher scores indicating higher anxiety (Spitzer, Kroenke, Williams, & Lowe, 2006.) The Teleconferencing group of n1 = 35 clients had a mean of M1 = 6.35 and a standard deviation of SD1 = 1.75. The Traditional face-to-face therapy group of n2 = 35 clients had a mean of M2 = 6.15 and a standard deviation SD2 = 1.81. Identify the null hypothesis to test whether the mean anxiety scores differed between the two unrelated groups.
u1 = u2
Which of the following represents the null hypothesis when doing a two-tailed test?
u1 = u2
What is the symbolic null hypothesis (H0) for this study?
u1 = u2 because in independent samples u1 will = u2 for the null instead of 0
Identify the research hypothesis to test whether the mean anxiety scores differed between the two unrelated groups.
u1 =/ u2
What is the symbolic research hypothesis (H1) for this study?
u1 =/ u2 basically the opposite of the null
Which of the following could represent a null hypothesis when doing a one-tailed independent measures t test?
u1 >/ u2
Levene's test is automatically run by SPSS to determine if the __________ of the two groups are significantly different.
variances
You should use an independent samples t test when __________.
you wish to compare the difference between two unrelated sample means