Lecture 5: Statistical testing

Question 1

The test to be used depends on?

Answer 1

The type of research question being asked and the design of the research

Question 2

What are Descriptive studies?

Answer 2

Designed to explore prevalence

May not include any complex statistical test

Question 3

The choice of the analysis depends on ?

Answer 3

the type of variables and levels of measurement used in the research

Question 4

What are levels of measurement?

Answer 4

Nominal
Ordinal
Interval/ratio

Question 5

What are Parametric tests?

Answer 5

Parametric tests make a number of assumptions about the distribution of the population from which the same is drawn ( e.g. normally distributed scores) and the nature of the data (interval or ratio level)

Question 6

What are Non-parametric tests?

Answer 6

Non-parametric tests do not make assumptions about the distribution of the data and are often more suitable techniques for smaller samples or for when the data collected is measure on the ordinal or nominal level

Question 7

What are two weaknesses of the non-parametric tests?

Answer 7

Less powerful - than parametric tests
Less likely to detect true differences or relationships that exist

Difficulty with interpretation - uses ranks rather than the raw value

Question 8

Type of ParMetric tests?

Answer 8

Paired sample T-test

Independent T-test

1- way ANOVA

2- way ANOVA

Pearson’s correlation

Question 9

Types of Non-parametric tests

Answer 9

Chi squared test

Wilcoxon signed- rank test

Mann-whitney U test

Kruskal- Wallis

Friedman

Spearman’s correlation

Question 10

What is Correlation ?

Answer 10

Correlation is measure of strength and direction of the association between two variables

Question 11

What is Pearson’s correlation?

Answer 11

Strength between two interval/ratio-level variables

Question 12

What is Spearman rank correlation (Spearman’s rho)?

Answer 12

compares two ordinal-level variables

Question 13

What is Point biserial?

Answer 13

correlation between categorical and continuous data

Question 14

What is a correlation coefficient : r?

Answer 14

A correlation coefficient (r) is a way to put a value to the relationship.

r ranges from -1 to +1

Question 15

What r (correlation coefficient)=0?

Answer 15

There is no relationship between the variables

Random scatter

Question 16

What r (correlation coefficient)= -1?

Answer 16

Perfect negative correlation

Question 17

What r (correlation coefficient)=+1?

Answer 17

Perfect positive correlation

Question 18

What statistical test to use to compare means for 2 groups ?

Answer 18

Two-sample T test = Student’s T test

Question 19

What statistical test to use to compare means for more than 2 Groups?

Answer 19

Analysis of variance= ANOVA

Question 20

What Statistical test to use to compare means for 1 Group?

Answer 20

Paired T test

Question 21

What Statistical test to use to compare proportions in a Big sample size?

Answer 21

Chi-square test

Question 22

What Statistical test to use to compare proportions in a small sample size?

Answer 22

Fisher’s exact test

Question 23

When do we use Independent sample t-test?

Answer 23

When we have two samples that are independent of each other and we want to compare the means of the two samples
•For example, you want to compare cholesterol levels between runners and swimmers.

Question 24

When do you use Paired sample t-test

Answer 24

When we have two samples that are NOT independent of each other and we want to compare the means.

•For example, you want to know if a certain diet is effective at reducing LDL cholesterol levels (the bad kind of cholesterol). Could apply to pre- and post-data. E.g. mean cholesterol levels pre- and post- starting a diet plan

Question 25

P value interpretation in Independent samples T-test

Answer 25

If the p-value < 0.05, there is a significant difference between the means of the two groups. Report the means of the two groups or the mean difference and confidence interval

Question 26

P value interpretation in a paired samples- T test?

Answer 26

If the p-values <0.05 then there is a statistically significant difference between the two time points/experiments. Report the mean difference.

Question 27

P value interpretation in Chi-squared test.

Answer 27

If p < 0.05, there is significant evidence that the proportions are not the same. Use %’s to describe what the relationship is.

Question 28

Dependent and independent variables in Independent samples T-test?

Answer 28

Dependent variable: Continuous Independent variable: Categorical

Question 29

Dependent and independent variables in Paired samples T-test?

Answer 29

Dependent variable: Continuous Time point: 1 or 2/ condition

Question 30

Dependent and independent variables in Chi-square test?

Answer 30

Dependent variable: Categorical

Independent variable: Categorical

Question 31

A study looks to determine the average cholesterol levels in smokers compared to non-smokers. Which statistical method should they use?

A Paired t Test
B Two sample t test
C Fishers exact test
D Pearson’s Chi Squared
E ANOVA
F Spearman’s Correlation Coefficient

Answer 31

B

Question 32

Researchers look to determine if there is a relationship between serum C reactive Protein levels and hormone replacement therapy.

Which is the best method to assess if HRT is related to CRP?
A  Paired t Test
B  Two Sample t Test
C. Fisher’s Exact
D. Chi-squared test

Answer 32

D

Question 33

An obese population is given a new weight loss drug for a year to determine its efficacy. The patients are evaluated at baseline and then one year later. Which test should they use to determine if this drug is effective?

A Paired t Test
B Two sample t test
C Fishers exact test
D Pearson’s Chi Squared
E ANOVA
F Spearman’s Correlation Coefficient

Answer 33

A