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
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Causal and Correlational Methodologies
Transcript of Causal and Correlational Methodologies
Boaler, J. (2014). Research suggests that timed tests cause math anxiety.
Teaching Children Mathematics
, 20 (8), 469-478, http://eds.b.ebscohost.com.proxy.kennesaw.edu/eds/detail?sid=000cfa44-7189-4f28-ab5d-6a02e86924e6%40sessionmgr115&vid=1&hid=101&bdata=JnNpdGU9ZWRzLWxpdmUmc2NvcGU9c2l0ZQ%3d%3d#db=edsgao&AN=edsgcl.364991609
Causal-comparative (ex-post facto) research. (n.d).
Area Education Agency
Correlation does not imply causation. (2014). In
Gokalp, M. S. (2013) Perceptions of the internet and education: A study with physics education website users.
International Journal of Environmental and Science Education
, 8 (2), 289-302. http://eds.b.ebscohost.com.proxy.kennesaw.edu/eds/detail?sid=56caedf0-cd96-4933-8391-309908241eda%40sessionmgr113&vid=1&hid=101&bdata=JnNpdGU9ZWRzLWxpdmUmc2NvcGU9c2l0ZQ%3d%3d#db=eric&AN=EJ1008606
Haynes, J. C., Ronbinson, J. S., Edwards, M. C., & Key, J. P. (2012). Assessing the effect of using a science-enhanced curriculum to improve agricultural students' science scores: A causal comparative study.
Journal of Agricultural Education
, 53 (2), 15-27. http://files.eric.ed.gov/fulltext/EJ993265.pdf
Park, H., Berhman, J. R., & Choi, J. (2013). Causal effects of single-sex schools on college entrance exams and college attendance: Random assignment in Seoul high schools.
, 50 (2), 447-469. http://eds.b.ebscohost.com.proxy.kennesaw.edu/eds/detail?sid=4bb1bba7-f576-4eaf-9125-a3edc883456e%40sessionmgr113&vid=1&hid=101&bdata=JnNpdGU9ZWRzLWxpdmUmc2NvcGU9c2l0ZQ%3d%3d#db=eoh&AN=1360660
Peters, J. (2013). When ice cream sales rise, so do homicides. Coincidence, or will your next cone murder you? Crime: A blog about murder, theft, and other wickedness. In
Sterns, E., Moller, S., Potochnik, S., & Blau, J. (2007). Staying back and dropping out: The relationship between school retention and school drop out.
Sociology of Education
, 80 (3), 210-240. http://proxy.kennesaw.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=sih&AN=26319527&site=eds-live&scope=site
Causation or correlation?
HDL cholesterol is higher in patients who have not had heart attacks and lower in those who have.
A study found that as ice cream consumption increased in Chicago, violent crime increased as well.
A study by the University of Pennsylvania found that people who slept with the light on when they were children were more likely to develop myopia later in life.
Atmospheric CO2 levels have been increasing since the 1950's; the obesity rate has risen dramatically in that same time period.
A study by Northwestern earlier this year found that college students who smoked marijuana had abnormalities in two regions of the brain. The more often they smoked, the greater the abnormality.
Types of causal relationships:
Looks for and studies patterns of variation in two or more variables
Not experimental, often observational - no manipulation by researchers
Frequently exploratory research to identify variables; may be followed up with causal-comparative research and/or experimental research
Seeks to determine which factors cause which effects
Also not experimental research
However, it has a similarly structured hypothesis and includes independent and dependent variables
Naturally comprised of two groups
Often studies variables that would be unethical to manipulate (e.g. feral children or less obviously unethical grade retention)
In order to prove causation, you need:
A relationship between variable A and B
A to precede B
*No other factors that could have caused B*
Limits of Correlational Research
Correlation may be due to:
A causing B (or A causing B to cause C)
B causing A (directly or indirectly)
A and B being caused by the same unknown factor
A causing B, which in turn causes A
P value shows likelihood of correlation being a coincidence. Correlations with p values ranging from .01 (1%) to .05 (5%) are considered "significant."
Does not show causation. Further study needed to reveal probable causation.
Types of causal research
Retrospective Causal-Comparative Research
Much more common
Looks back, tries to determine if a factor caused an event or events that have already happened.
Prospective Causal-Comparative Research
Attempts to determine cause and effect nature beginning with a factor and observing its effects.
Limitations of causal-comparative research
Researchers need to be careful to select groups that only differ by the independent variable identified in their study.
Problems with extraneous variables can be mitigated by:
Finding very similar groups
Dividing data into sub groups
Analysis of covariance - a statistical procedure
Researchers can not decide how to divide groups.
For complex relationships, causation can not be proven, even in a causal-comparative study. Studies should label discoveries as "possible cause" and "possible effect" (
Causal-comparative (ex-post facto) research)
Often used to determine if an experimental study is worthwhile or because an experimental study would be unethical.
Examples of causal comparative research
When factor A increases, effect B also increases. When A decreases, B decreases.
Scaled from 0 to +1.0. Zero shows no correlation, +1 shows complete, positive correlation.
When factor A increases, effect B decreases. When A decreases, B increases.
Measured from 0 to -1.0. Zero shows no correlation, -1 shows complete negative correlation.
Is this a good example of a casual-comparative research study? (Or is it more correlational?)
Is it a retrospective or prospective study?
What is the hypothesis?
What are the independent and dependent variables?
How do they try to minimize extraneous variables?
What are the results?
What criticisms (if any) can you make of this study?
A 2014 study found that timed math tests cause text anxiety to begin early in a student's live (Boaler, J).
A 2013 study which found that a person gender, occupation, and the amount of time they spent on the internet were all factors affecting how they viewed physics education websites (Gokalp, M. S).
A Seoul-based study found that attending same-sex schools led to higher grades on college entrance exams (Park, Berhman, & Choi, 2013).
A study that finds that students held back in eighth grade are much more likely to drop out in high school and argues that dropping out was an effect of retention (Sterns, Moller, Potochnik, & Blau).
(Correlation does not imply causation)
Don't like the terms causal or correlational? Neither does:
Johnson, B. (2000). It's (beyond) time to drop the terms causal-comparative and correlational research in educational research methods textbooks. http://proxy.kennesaw.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=eric&AN=ED445010&site=eds-live&scope=site