68 Questions
| Total Attempts: 226

This is for statistics. It covers topics, including ANOVA, t-tests, and definitions.

Questions and Answers

- 1.A dichotomous variable should be labeled as ____________ for quantitative data analysis.
- A.
Nominal

- B.
Ordinal

- C.
Interval

- D.
Ratio

- 2.If a variable is measured in equal intervals and has a true zero, it should be labeled as ____________ for quantitative data analysis.
- A.
Nominal

- B.
Ordinal

- C.
Interval

- D.
Ratio

- 3.If a researcher selects every 10th person, for a total of 100 subjects, from a list of 1,000 available subjects, the investigator has performed ________________ sampling.
- A.
Random

- B.
Stratified

- C.
Systematic

- D.
Cluster

- 4.The equation (standard deviation/mean) is called the standard error of the mean and is used in the calculation of a 95% confidence interval.
- A.
True

- B.
False

- 5.A stem and leaf diagram is very useful when data is measured on an ordinal scale.
- A.
True

- B.
False

- 6.The JMP software package uses the t-distribution to calculate confidence intervals rather than the z-distribution when the sample size is small.
- A.
True

- B.
False

- 7.6. A Shapiro-Wilk test that has a p-value of 0.03 would indicate that the variable of interest is not normally distributed.
- A.
True

- B.
False

- 8.A distribution that is severely skewed to the right will have a mean that is larger than the median
- A.
True

- B.
False

- 9.As the sample size goes up and the variance goes down, the 95% confidence interval becomes tighter (smaller).
- A.
True

- B.
False

- 10.Each of the following statistical tests have an assumption of independence: Independent t-Test, Paired t-Test, Completely Random (CRD) ANOVA.
- A.
True

- B.
False

- 11.Dr. John is conducting a study to predict the probability of being male or female (Y) based on the number of sarcomeres identified from a muscles biopsy (X). Dr. John should us e a Chi-square analysis for this investigation.
- A.
True

- B.
False

- 12.To increase statistical power, you should decrease your alpha and increase your sample size.
- A.
True

- B.
False

- 13.Power is related to all of the following except one, which one is it?
- A.
Variance

- B.
Alpha level

- C.
Sample Size

- D.
Effect Size

- E.
SEM

- 14.In which part of the research paper should you talk about your limitations and weaknesses?
- A.
Introduction

- B.
Abstract

- C.
Discussion

- D.
Methods

- 15.When performing a statistical analysis, if your sample size is larger than 30, what can you use to defend your data?
- A.
Bonferroni Correction

- B.
Central Limit Theorem

- C.
Beta

- D.
Type II error

- 16.ICC is used to measure _________; PPMCC is used to measure ___________.
- A.
Correlation, Agreement

- B.
Agreement, Power

- C.
Agreement, Correlation

- D.
Correlation, Power

- 17.What is the null hypothesis?
- A.
The hypothesis that there is a difference.

- B.
The hypothesis that there is no change.

- 18.We will be confident a true difference exits when:
- A.
We have a large error and a larger treatment effect

- B.
We have a small error and small treatment effect

- C.
We have a large treatment effect and small error

- 19.Which of the following is not true about a confidence interval?
- A.
As SEM decreases, the CI decreases

- B.
It is an interval Estimate

- C.
The bigger the N, the wider the CI

- D.
The larger the alpha level, the wider the CI

- E.
The smaller the n, The wider the CI

- 20.If we have a small sample size, which distribution should we use?
- A.
T

- B.
Z

- C.
F

- D.
P

- 21._____ + s/ square root of n x 100
- A.
Coefficeient of variance

- B.
Confidence Interval

- C.
SEM

- D.
Variance

- 22.Which of the following statements is true?
- A.
Standard deviation is the variance squared.

- B.
Variance is standard deviation squared.

- C.
We are most interested in the variance.

- 23.Grading a MMT is an example of which type of scale:
- A.
Ordinal

- B.
Nominal

- C.
Interval

- D.
Ratio

- 24.Which of the following is not a parametric assumption?
- A.
Normality

- B.
Equal variance

- C.
Independence

- D.
Random Assignment

- E.
Unequal varaince

- 25.Mean, Mode, and median are all measures of central tendency and when a distribution is skewed, we should use the median.
- A.
True

- B.
False

- 26.In which section of a research paper should we report our findings?
- A.
Intro

- B.
Abstract

- C.
Discussion

- D.
Methods

- E.
Results

- 27.The varaible that we manipulate is called the:
- A.
Dependent

- B.
Independent

- 28.What is one of the major assumptions crossing every statistic?
- A.
Random

- B.
Independent

- C.
Equal Variances

- 29.Which of the following is the highest level of evidence?
- A.
Case-control

- B.
Cohort study

- C.
Individual case-control

- D.
Case series

- E.
Outcomes research

- 30.We can make predictions using descriptive statistics.
- A.
True

- B.
False

- 31.What does kurtosis measure?
- A.
Roundness

- B.
Peakedness

- 32.To increase power, we should do all of the following except one:
- A.
Increase n

- B.
Increase variance

- C.
Decrease variance

- D.
Increase effect size

- E.
Increase alpha

- 33.When we try to separate extraneous variables to uncover the true nature of how one phenomenon relates to another, it is said to be:
- A.
Systematic

- B.
Empirical

- C.
Conrolled

- D.
Critical

- 34.Objective Data based on direct observation:
- A.
Empirical

- B.
Systematic

- C.
Controlled

- D.
Critical

- 35.When we have a Nominal Y and a Nominal X, which type of test should we do?
- A.
Chi square

- B.
T-test

- C.
Logistic Regression

- D.
Correlation and Regression

- 36.When should we do a correlation or regression analysis?
- A.
Nominal X, Continuous Y

- B.
Continous Y and X

- C.
Nominal Y and X

- 37.What is the only nonparametric assumption?
- A.
Random

- B.
Equal Variances

- C.
Independent

- 38.What is true of a paired t-test?
- A.
YOu can use it to make predictions.

- B.
It is robust to parametric assumptions.

- C.
You can use it when you have 4 variables.

- 39.When we fail to reject a false null, what type of error is that?
- A.
Type I

- B.
Type II

- 40.We minimize a Type I error by __________; a Type II error by _________.
- A.
Large n, a low alpha

- B.
A low alpha, a small n

- C.
A low alpha, a large n

- D.
A high alpha, a large n

- 41.An effect size of .5 is considered to be a large effect size and that is what we want.
- A.
True

- B.
False

- 42.What is the proper name for a post hoc?
- A.
Multiple Comparisons Procedure

- B.
Multiple Regression

- C.
Standard Error of the Estimate

- D.
Variance

- 43.What does the R^2 tell you?
- A.
That ___% of the variability in the dependent is explained by the independent variables used.

- B.
That ___% of the variability in the independent is explained by the dependent variables used.

- 44.What type of ANOVA is an extension of an RCB?
- A.
Latin Square

- B.
CRD

- C.
Factorial

- D.
Nested

- 45.Dr. X is examining has data about Age and Gender on Sprint Time. Prior investigation has shown that Gender does make a difference. In oder to examine this data, what type of analysis should we complete? How do we do that in JMP?
- A.
Independent T-test, Fit Y By X

- B.
Randomized Complete Block, Fit Y by X

- C.
Nested, Fit Model

- D.
Factorial Anova, Fit Model

- 46.We cannot do a CRD if there is not:
- A.
A high alpha

- B.
Independence

- C.
Random assignment

- 47.You conduct a study to examine the effects of joint mobilization on patients with adhesive capsulitis. Your design has four groups (group A gets heat prior to mobilization, group B gets ice after mobilization, group C gets heat prior and ice after mobilization, group D is the control group and does not receive a treatment. The results after 2 weeks of treatment reveals the following improvements in shoulder ROM. What is the appropriate analysis?
- A.
Factorial

- B.
RCB

- C.
Latin Square

- D.
CRD

- 48.You conduct a study to examine the relationship between age and the number of occurrences of back pain that a person has experienced in their lifetime. You sample 10 people at random from the phone book. What statistical test can be used to predict the number of occurrences of back pain that someone should expect to have by the time they are 40 years old?
- A.
Correlation

- B.
CRD ANOVA

- C.
Independent t-test

- D.
Regression

- 49.Which statement is true about a factorial ANOVA?
- A.
If there is an interaction effect, look at the main effects.

- B.
IF there is an interaction effect, look at the main effects.

- C.
YOu should look at the main effects first.

- D.
If there is an interaction effect, you do not need to look at the main effects.

- 50.____ statistics estimate _________ parameters.
- A.
Population, Sample

- B.
Descriptive, Population

- C.
Sample, Population

- 51.You can use a t-test in place of an ANOVA.
- A.
True

- B.
False

- 52.When do we use a CRD?
- A.
When there are more than 3 independent variables.

- B.
When there are more than 2 levels of the independent variable.

- C.
When there are less than 2 levels of the independent variable.

- 53.Why can a t-test not handle more than 2 groups?
- A.
Too much variance.

- B.
Decreased power.

- C.
Inflation of alpha.

- D.
Too few subjects.

- 54.What is the F test?
- A.
Variance between / variance within

- B.
Variance within / variance between

- 55.Which of the following is not true of a Repeated Measures ANOVA?
- A.
Logical extension of a paired t

- B.
Measures are independent

- C.
Measures are on the same subject over time

- D.
You do a MANOVA

- 56.What is the residual really a measure of? Do we want it large or small?
- A.
Error, small

- B.
Error, large

- C.
Effect, small

- D.
Effect, large

- 57.In a randomized complete block, the blocking variable is also known as the __________. This design attempts to increase statistical power by ____________.
- A.
Regressor variable, decreased error term

- B.
Nuisance, increased error term

- C.
Nuisance, decreased error term

- D.
Regressor, increased treatment effect

- 58.If you have agreement, you will have correlation.
- A.
True

- B.
False

- 59.Why is multicolinearity a problem?
- A.
It can completely change the sign of parameter estimates.

- B.
It means the data is not linear.

- C.
It is the reflection of high variance.

- 60.When testing for Multicolinearity, which of the following statements is true?
- A.
We want correlation

- B.
We want a correlation < .8, Eigon values close to one, a condition number between 5-10, and VIFs below 10.

- C.
We want Eigon values close to two, a large condition number.

- 61.Which of the following condition numbers reflects the most MC?
- A.
5-10.

- B.
30-100.

- C.
>1,000.

- 62.When we do a stepwise regression, we generally want our ______ model
- A.
Most complex.

- B.
Simplest.

- C.
One with the smallest R^2.

- 63.When you are determining a model, which of the following is not true?
- A.
You want a large R^2 and R^2 adjusted

- B.
You want CP = P

- C.
You want a large PRESS

- D.
You want a small PRESS

- E.
You want small MSE

- 64.You want a large within groups, small between groups.
- A.
TRUE

- B.
FALSE

- 65.The large the F, the greater the power, the greater the variability, the more likely to reject the null.
- A.
True

- B.
False

- 66.Regression analysis is used for all of the following but one:
- A.
To see if agreement exists

- B.
Model Building

- C.
Varaible screening

- D.
Parameter estimation

- E.
Prediction

- 67.What is the primary assumption for regression analysis?
- A.
IV is normal.

- B.
Residuals are normal.

- C.
DV is normal.

- 68.If we are doing correlation, a cigar shape represents what?
- A.
No correlation

- B.
Strong correlation

- C.
Weak correlation