SU PHE5020 2021 July Knowledge Checks Latest (Full)

PHE5020 Biostatistical Methods

Week 1 Knowledge Check  

Question 1 The statistical process of using samples to estimate population parameters is known as:

Statistical interference.

Statistical inference.

Statistical confidence.

Descriptive statistics.

Question 2 If a researcher wishes to determine whether there is evidence that the mean fasting glucose level in adults with type 2 diabetes is different from 110mg/dL, then

a two-dependent samples t-test should be considered.

a two-independent sample t-test should be considered.

a one-sample t-test should be considered.

either a one-sample or two-dependent samples t-test should be considered.

Question 3 Power is defined as:

the probability that you will retain/keep the null hypothesis if it is false.

the probability that you will reject the null hypothesis if it is true.

by your research hypothesis only.

the probability that you will reject the null hypothesis if it is false.

Question 4The probability of rejecting the null hypothesis that is true is known as the:

Type I error

Type II error

statistical power

confidence level

Question 5 Two random samples have sizes of n=100 and n=75 respectively. Which of the following is true for a 95% confidence interval?

The sample of n=75 has a greater degree of confidence.

The sample of n=100 has a greater degree of confidence.

The confidence interval for the sample of n=100 is narrower.

The confidence interval for the sample of n=100 is wider.

 

 

PHE5020 Biostatistical Methods

Week 2 Knowledge Check  

Question 1  Which of the following statements is not true of parametric statistics?

They are inferential tests.

They assume certain characteristics of population parameters.

They assume normality of the population.

They are distribution-free.

Question 2 Assume you are conducting a study and find that the data violate all the assumptions of the statistic you had planned to conduct, what are your alternatives?

Conduct the statistical test since there will be no evidence of the error.

Run the study again and hope the data are better.

Conduct a nonparametric statistical test.

Change the research question.

Question 3 Under what circumstances would you use a non- parametric test?

In a pilot study.

When your data does not meet the assumptions for a parametric test.

When you think your sample size is too big.

When you do not really understand a parametric test.

Question 4Nonparametric tests are just as powerful as their parametric counterparts.

True

False

Question 5The Mann-Whitney U test is the nonparametric counterpart for which parametric test?

One-sample t-test.

Two-dependent sample t-test

Two-independent sample t-test.

Kruskal-Wallis Test

 

 

 

 

PHE5020 Biostatistical Methods

Week 3 Knowledge Check  

Question 1: Chi-square test for independence assesses which of the following?

It assesses whether there is a relationship between the population and the sample.

It assesses whether there is a relationship between two categorical variables.

It assesses whether there is significant difference between scores taken at time 1 and those taken at time 2.

It assesses whether the minimum number of cases exceeds recommended boundaries.

Question 2: Which tests could be used if in a contingency table your expected cases were fewer than what is required for the Chi-square test?

Chi-square test for independence.

Two-independent sample t-test.

Fisher's Exact Probability Test.

Paired-samples t-tests.

Question 3: Contingency tables and degrees of freedom are key elements of the chi-square test.

Question 3 options:

True

False

Question 4: For the chi-square test to be effective, the expected value for each cell in the contingency table has to be at least:

1

5

10

the number of rows times the number of columns.

Question 5: When an odds ratio is calculated from a 2x2 table:

The odds ratio is a measure of the strength of the relationship between the risk factor and disease variables.

The ratio may take any positive or negative value.

An odds ratio of 5 indicates there is no increase in odds of disease among the exposed group.

An odds ratio of 0.50 indicates there is an increase in odds of disease among the exposed group.

 

 

 

PHE5020 Biostatistical Methods

Week 4 Knowledge Check  

Question 1: Correlation refers to:

The causal relationship between two variables.

The linear relationship between two variables.

The proportion of variance that two variables share.

A statistical method that can only be used with a correlational research design.

Question 2: If two variables are highly correlated, what do you know?

That they always go together.

That high values on one variable lead to high values on the other variable.

That there are no other variables responsible for the relationship.

That changes in one variable are accompanied by predictable changes in the other.

Question 3: The coefficient of determination tells us:

The proportion of variance in X accounted for by the mean of Y.

The proportion of variance in Y accounted for by X.

The mean value of Y.

The mean value of X.

Question 4: In the equation of a straight line, Y = mX + c the term, m is the:

Independent variable.

Dependent variable.

Intercept.

Slope.

Question 5: Analysis of Variance (ANOVA) is a test for equality of:

variances.

means.

proportions.

only two parameters

 

 

 

PHE5020 Biostatistical Methods

Week 5 Knowledge Check  

Question 1 All the following statements are true for the odds ratio except:

can be calculated obtained from case-control studies.

 it is the incidence between the exposed divided the incidence between the un-exposed.

 it is an estimate of relative risk.

 it tends to be biased towards 1.

Question 2 : The analysis of epidemiological studies is based on the following except:

 contingency tables

 statistics that measure effects

 measuring the confounders

 experimental effect size.

Question 3  Epidemiology describes the distribution of disease in terms of Person, Place and Time.

True

 False

Question 4 Epidemiological studies usually describe the characteristics of people in terms of age, gender, race, ethnicity, socioeconomic status and religion, among others.

True

 False

Question 5  The most important difference between odds ratios and relative risks is that odds ratios measure incidence directly from the observed data.

True

 False