8. MEAN DIFFERENCE Flashcards
1
Q
- In which situations is the Mean Difference most used?
A
- when the exposure is categorical
and the outcome is numeric
2
Q
- Fill in the labels for the blocks numbered 1-6.
A
- Binary Outcomes
- Existing Cases
- New Cases
- Odds Ratio
- Risk Ratio
- Rate Ratio
3
Q
- Fill in the labels for the blocks numbered 7-12.
A
- Numeric Outcomes
- Categorical Exposure
- Numeric Exposure
- Mean Difference
- Correlation Coefficient
- Regression Coefficient
4
Q
- What Measure of Association do you use when you have a Categorical Exposure?
A
- Mean Difference
5
Q
- What two Measures of Association do you use when you have Numeric Exposures?
A
- Correlation Coefficient
- Regression Coefficient
6
Q
- What does the Mean Difference refer to?
A
- it refers to the comparison between two means
- this is the comparison between two averages
7
Q
- Describe Mean Difference.
A
- it is a measure of Association
- it assesses the presence of an association between a
categorical exposure and a numeric outcome
8
Q
- What are the 2 Main Research Situations where the Mean Difference is applicable?
A
- COMPARING MEANS BETWEEN CATEGORIES
(GROUPS) OF CATEGORICAL VARIABLES- this is known as a Between Subjects Design
- EG: a comparison between independent groups of
individuals
- COMPARING MEANS IN A SINGLE GROUP OF
INDIVIDUALS IN TWO OR MORE DIFFERENT TIME
POINTS- this is known as a Within Subjects Design
- it can also be called Repeated Measures
- EG: a comparison before and after an intervention
9
Q
- How do we Mathematically calculate the Mean Difference between two independent groups?
A
10
Q
- Assuming that you are making use of Normally Distributed Variables that are independent from one another, what statistical tests would you use?
A
- INDEPENDENT SAMPLES T-TEST
- this is used when you have 2 categories
- it is used when you are comparing between subjects
- it is used to compare the mean difference between 2
independent variables
- ANALYSIS OF VARIANCE (ANOVA)
- this is used when you have more than 2 categories
- it is used when you are comparing between subjects
- it is used to compare the mean difference between
more than 2 independent variables
11
Q
- How do we mathematically calculate the Mean Difference between two time points surrounding a single group?
A
12
Q
- Assuming that you are making use of Normally Distributed Variables that are centred around a singular group and its paired samples, what statistical tests would you use?
A
- PAIRED SAMPLES T-TEST
- this is used when you have 2 different outcomes
surrounding the same exposure - it is used when you are comparing within subjects
- this is used when you have 2 different outcomes
- REPEATED MEASURES ANALYSIS OF VARIANCE
(REPEATED MEASURES ANOVA)- this is used when you have more than 2 different
outcomes surrounding the same exposure - it is used when you are comparing within subjects
- this is used when you have more than 2 different
13
Q
- What are Within Subjects Designs?
(Repeated Measures Designs)
A
- they compare the measurements in the same
participants - the outcomes come from the same sources of
variability
14
Q
- How are Repeated Measurements collected?
A
- they can be collected at different time points
- this would asses the change over time
15
Q
- Other than comparing one group between different time periods, how else can they be compared when it comes to Within-Subjects comparisons?
A
- the same participants could be compared under two or
more different conditions
EXAMPLE:
- comparing the pain intensity for the same participants
receiving different pharmacotherapies for pain relief
16
Q
- What is the general rule when it comes to dealing with the same participants?
A
- this is a Within Subjects Design
- the same participants are being compared
- we will make use of a Paired T-Test or a Repeated
Measures ANOVA
17
Q
- What is a more complicated type of Mean Difference Research Design that we can get?
A
- some research designs can combine both Between-
Subject tests and Within-Subject tests - we may have 2 or more groups
- this is a Between-Subjects Factor
- these 2 or more groups could be analysed over 2 or
more time points - this is a Within-Subjects Factor
18
Q
- What Statistical Test would we use in the more complicated Mean Difference Research Design?
A
- MIXED DESIGN ANALYSIS OF VARIANCE
(MIXED DESIGN ANOVA)- this is a statistical test that provides a mean
difference between two or more independent groups
(Between Subjects IV)
- this is a statistical test that provides a mean
- this test simultaneously subjects participants to
repeated measures of a numeric outcome (DV)
(Within Subjects IV)