Year 2 Research Methods Flashcards
When should a histogram be used?
When there is continuous data (i.e. height, weight, time / when one axis represents frequency).
When should a bar chart be used?
When data is represented in two or more categories (e.g. males/females).
The bars on a bar chart must be separate!!
How is a sign test conducted?
1) . Work out the difference between the two sets of data (+ or - (= if they are the same)).
2) . Add the number of positives and negatives.
3) . Calculated value / S value = least frequent sign.
4) . N value = total number of positive and negative signs (ignore ‘=’ signs).
5) . Compare the S value with the critical value (must be less than or equal to critical value to be significant).
What is a type I error?
When a researcher incorrectly rejects a true null hypothesis. E.g. findings are reported as significant thought they actually occurred by chance.
What are the requirements for a bar chart?
- Bars must not be joined
- Axis must be operationalised when labelled (e.g. number of metres ran)
- Number on the y-axis must be at regular intervals
- The chart must have a title.
- independent variable must be x-axis, dependent variable must be y-axis.
How are histograms constucted?
The frequency density must be calculated. Then the width of the bar is proportional to the class width. Bars should touch each other.
How is frequency density calculated?
Class width (difference between two numbers) / frequency.
What should the y-axis be labelled as on a histogram?
Frequency density.
What is a type II error?
Researcher fails to reject a null hypothesis which is actually false. E.g. the findings are reported as insignificant when they actually are.
How is percentage decrease calculated?
Previous value - new value x 100
Outline the process of peer review.
The report of the study is distributed to several established psychologist who will review and report back areas of amendment.
What is a single-blind review?
The names of the reviewers are not given to the researcher - prevents interference by the researcher.
What is a double-blind review?
Both the names of the reviewers and names of the researchers are not known by either party - ensures biases on the basis of gender, ethnicity and so on do not occur.
What is an open review?
The reviewers and researchers are known to each other.
What is a content analysis?
A method of quantifying qualitative data through coding units. E.g. number of slang words used.
Give an advantage of content analysis.
Advantages of Content Analysis:
- Reliable - content analysis is easy to replicate as coding units can be re-used.
- Compliments other methodology - supports longitudinal studies as coding units may be used over time.
Give a disadvantage of content analysis.
Disadvantages of content analysis:
- Lack of causation - not performed under controlled conditions therefore cannot be used to show causation.
What is counterbalancing?
The process of reversing the study with one half of a sample.
Where can counterbalancing be used?
Repeated measures design studies only.
What is the purpose of counterbalancing?
To reduce order effects.
List the 12 features of a scientific report in the order they should appear.
1) . Title
2) . Table of Contents
3) . Abstract
4) . Introduction
5) . Aims
6) . Hypothesis
7) . Procedure/method
8) . Findings
9) . Discussion
10) . Conclusion
11) . References
12) . Appendices
Define nominal data.
Data that can only appear in one category, once. E.g. naming you favourite football team.
Define ordinal data.
Numerical data though intervals are subjective - e.g. asking someone to rate something on a scale of 1-10.
Define interval data.
Data that is numerical on an ordered scale - e.g. tally scores (8 is twice as much as 4).
What are the three tests that can be used in correlational studies?
Chi-Squared
Spearman’s rho
Pearson’s r
What test is used in a correlational study with nominal data?
Chi-Squared.
What test is used in a correlational study with ordinal data?
Spearman’s rho.
What test is used in a correlational study with interval data?
Pearson’s r.
Name the three tests that can be used in a study of difference with a related design.
Sign test
Wilcoxon
Related t-test
What test is to be used when the study of difference has a related design and nominal data?
Sign test.
What test is to be used when the study of difference has a related design and ordinal data?
Wilcoxon.
Which test should be used when the study of difference has a related design and interval data?
Related t-test.
Name the three tests that could be used in a study of difference with an unrelated design.
Chi-squared
Mann-Whitney
Unrelated t-test.
Which test should be used when the study of difference has an unrelated design and nominal data?
Chi-squared
Which test should be used when a study of difference has an unrelated design and ordinal data?
Mann-Whitney.
Which test should be used when a test of difference has an unrelated design and interval data?
Unrelated t-test.
Define paradigm shift.
An important change over time in the basic principles and experimental practices of a scientific discipline.
What is a thematic analysis?
When the researcher will repeatedly review their study to identify themes to quantify the qualitative data.
