QUESTION 1
For which of the following correlations would the data points be clustered most closely around the regression line?
a) r= 0.00
b) r= 0.05
c) r= -0.83
d) r= 0.79
e) There is no relationship between the correlation and how close the points are to the regression line.
QUESTION 2
The sample data for a chi-square test are called______________
a) observed proportions
b) expected proportions
c) observed frequencies
d) expected frequencies
QUESTION 3
A researcher was comparing several exercise groups of female subjects to examine the effect on mean waist circumference change following 12 weeks of exercise. The researcher obtained an F-ratio of 5.27 with df = 3, 37 from an independent-measures One Way ANOVA. Assuming there are no missing values, what is the total number of treatment groups G being compared in the study?
a) There is insufficient information to calculate the answer
b) G = 41
c) G = 40
d) G = 4
e) G = 3
QUESTION 4
A researcher was comparing several exercise groups of females to examine the effect on mean waist circumference change following 12 weeks of exercise. The researcher obtained an F-ratio of 5.27 with df = 3, 37 from an independent-measures One Way ANOVA. Assuming there are no missing values, what is the total number of female subjects N employed in the study?
a) N = 41
b) N = 3
c) N = 40
d) N = 4
e) There is insufficient information to calculate the answer
QUESTION 5
A researcher was comparing several exercise groups of female subjects to examine the effect on mean waist circumference change following 12 weeks of exercise. The researcher obtained an F-ratio of 5.27 with df = 3,37 from an independent-measures One Way ANOVA. What p-value should the researcher report when she writes up the analysis for a good quality journal?
a) p<0.001
b) p=0.003966135
c) p=0.004
d) p=5.27
e) There is insufficient information to calculate the answer
f) p=0.003
QUESTION 6
A hypothesis test is ________________
a) an inferential technique that uses information about a population to make predictions about a sample
b) a descriptive technique that allows researchers to describe a sample
c) a statistical technique that always compares two or more means
d) an inferential technique that uses the data from a sample to draw inferences about a population
QUESTION 7
The two-by-two table below examines the relationship between receiving a drug and it’s side effects. The cells show a number of observed and expected results. Three observed and three expected values are missing. Fill in the missing values.
Report your values to ONE DECIMAL PLACE in the format XX.X.
QUESTION 8
Using the information from the previous questions, a chi squared test was run to determine the association between the drug and side-effects. What is the test statistic rounded to TWO DECIMAL PLACES X.XX?
QUESTION 9
What is the number of degrees of freedom for this chi squared statistic? Report your answer as a whole number.
QUESTION 10
What is the p-value for this chi squared statistic? Report your answer to THREE DECIMAL PLACES X.XXX.
QUESTION 11
Using the Chi Squared test above for drug by side-effects and a significance level of 0.05, does this p-value lead us to accept or reject the null hypothesis?
a) Reject null hypothesis, there is insufficient evidence of a relationship between drug and side-effects
b) Reject null hypothesis, there is evidence of a relationship between drug and side-effects
c) Accept null hypothesis, there is insufficient evidence of a relationship between drug and side-effects
d) Accept null hypothesis, there is a relationship between drug and side-effects
QUESTION 12
Based on this evidence from this study alone, is the drug free of side-effects?
a) No, there was no statistically significant difference in the proportions of side effects experienced by subjects on the drug.
b) No, a significantly higher proportion of subjects had side effects on the drug than expected.
c) Yes, there was no statistically significant difference in the proportions of side effects experienced by subjects on the drug.
d) Yes, a significantly higher proportion of subjects had side effects on the drug than expected.
QUESTION 13
Which of the following is a characteristic of a nonparametric test?
a) A numerical score is required for each individual.
b) The hypotheses concern population means and variances.
c) The test requires assumptions about the population means or variances.
d) None of the above
QUESTION 14
A researcher is examining the relationship between pill packaging preferences and gender. A sample of 27 men and 27 women is obtained and each person is asked to identify his/her preference between two choices of packaging for a new malaria tablet. For this sample, 9 of the men preferred packaging A, and 15 of the women preferred packaging A. If a chi-square test is used to evaluate the relationship, what is the expected frequency for women preferring packaging A?
a) 9
b) 12
c) 15
d) 27
QUESTION 15
The following data give the self-reported heights (repht) in cm and weights (repwt) in kg for a random sample of 111 women in a US study. The study was designed to look at the relationship between self reported weights and heights with those measured at a clinic. However, here we are interested in predicting height from weight using the self reported figures. The data are available in the text file PUBH7630_A2_htwt.csv (attached to order)
(A) Using the scatterplot above and data (available as a text file above), is a simple regression test justified? What else should be checked before (or after) you proceed with a simple linear regression? State your explanation in the space provided. (4 Marks)
If you are having trouble seeing the scatterplot it is also available as A2_scatterplot.pdf (attached to order)
(B) Carry out a linear regression, providing the equation and SEs of coeffcients along with your interpretation (4 marks).
QUESTION 16
Data were collected on 655 boys aged 10-15 from Tecumseh, Michigan which included height and forced expiratory volume (FEV). You are asked to analyse the relationship between the predictor variable, the boys height and the outcome variable their FEV using a simple linear regression.
PLEASE MATCH THE QUESTION ITEMS ON THE LEFT WITH THEIR RELEVANT CONCEPTS FROM THE ANSWER ITEMS ON THE RIGHT. ANSWER ITEMS ON THE RIGHT CAN BE USED MORE THAN ONCE.
1) FEV
2) Height
3) Linearity
4) An expected number greater than 5 in each cell
5) Normality of the outcome variable
6) Normality of the predictor variable
7) Normality of the residuals
QUESTION 17
A chi-square test for independence is used to evaluate the relationship between two variables. If one variable is classified into 4 categories and the other variable is classified into 2 categories, then the chi-square statistic will have
a) df=8
b) cannot determine the value of df from the information provided.
c) df=3
d) df=6
QUESTION 18
1. Which of the following sets of data provides the clearest difference between the two samples assuming they have the same sample size? (Note that M = mean and s is the standard deviation).
a. One sample has M = 120 with s2 = 20 and the other has M = 130 with s2 = 20
b. One sample has M = 120 with s2 = 5 and the other has M = 125 with s2 = 5
c. One sample has M = 120 with s2 = 5 and the other has M = 130 with s2 = 5
d. One sample has M = 120 with s2 = 20 and the other has M = 125 with s2 = 20
QUESTION 19
A researcher is examining preferences among 3 brands of cat food using a sample of n = 60 dog owners each with one cat. Each owner allows their cat to taste all 3 brands and selects the one he or she perceives to be the cat’s favourite. If the data are evaluated with a chi-square test for goodness of fit using a = .05, then a significant result would require a chi-square statistic
a) less than 5.99
b) greater than 5.99
c) less than 7.81
d) greater than 7.81
