Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Research Methods Lecture 13 Quantitative Methods
Transcript of Research Methods Lecture 13 Quantitative Methods
Know what kind of numbers you have
Nominal, ordinal, interval
Statistical analysis can tell us if there's an association between variables
Whether knowing the value of one improves our odds of correctly guessing the corresponding value for the second
Coefficient of association: number that summarizes the association
0 to 1 or -1 to 1 with numbers closest to zero being a weak association
Direction of the association
Positive or negative
Higher/Higher v. Higher/Lower
Examples: -.87; .2
With a data analysis program like SPSS you can combine variables, assign control variables, and run multivariate regressions
If used, these analysis techniques should be fully explained in the article, either in the text, the notes, or the appendix
To Do Quantitative Analysis
Writing a Research Paper
Developing the thesis
Writing the paper
How do you define terms?
Quality of life?
And THEN: how do you quantify (assign numbers to) a variable that is not already a number?
How have other researchers quantified variables in the data sets you can access and use OR that are used in the research you are reading?
You have to understand THEIR methodology
WHAT do the numbers MEAN?
Yes, research is hard
Sometimes you have to spend hours looking for resources
Sometimes the perfect resource isn't available
BUT there are TONS of journals and databases through the VT library that you can access free of charge
Different methods for a position paper v. a research assignment
Dr. Courtney Thomas
Lecture 13: Quantitative Methods
Look for relationships between quantified variables
Categorize by assigning a number
Male = 1 Female = 2
Categorize and rank
Categorize, rank, distance
Validity and Reliability
Validity: Do the values assigned directly relate to or capture the concept we are trying to measure or understand?
Reliability: Are the values assigned consistently and are they related to each other in a meaningful way?
If you are going to do your own quantitative analysis you need additional coursework
STAT 2004: Intro to Stat
STAT 3604: Stat Social Sciences
Look for a course that specifically teaches SPSS or a similar program
Where Do You
Get Your Data?
Collect your own through surveys, interviews, focus groups
THOUSANDS of them
Available through organizations, libraries, governments, etc.
But you HAVE TO UNDERSTAND how
the data were collected and coded!
Articles, books, etc.
Where did they get their data? How?
How likely is it that the association we have measured between two variables actually exists?
90% confidence or statistically significant to .1
95% confidence or statistically significant to .05
99% confidence or statistically significant to .01
Measures of Association
Lambda: association (0 to 1)
Chi-Square: statistical significance
Gamma: association (-1 to 1 where 0 is the absence of association)
Tau-b: association with same number of categories
Tau-c: association with different number of categories
Standard score of gamma: significance
or correlation coefficient measures explained variance (-1 to 1)
r2 measures association
Can only determine significance when both variables are normally distributed