KL Ch. 20 General Designs of Research Flashcards Preview

ECPY 767 (Final) > KL Ch. 20 General Designs of Research > Flashcards

Flashcards in KL Ch. 20 General Designs of Research Deck (8)
Loading flashcards...

Map of Experimental Group-Control Group: Randomized Participants

R ​X ​Y ​(experimental)
R ​-X ​Y​ (non-experimental)


Experimental Group-Control Group: Randomized Participants controls for...

History, maturation, and presenting


Advantages of Experimental Group-Control Group: Randomized Participant

1) Has the best built-in theoretical control system of any design, with one or two special exceptions.
2) It is flexible, theoretically able to extend to any # of groups with any # of variables.
3) If extended to more than on variable, it can test several hypothesis at one time.
4) It is statistically and structurally elegant.


Map of Experimental Group- Group: Matched Participants

[Mr] ​ X​ Y ​(experimental)
[Mr] ​-X​ Y​ (control)


Matching Versus Randomization

- Sometimes matching is done in the field, when the researcher wants to remain discrete.
- Can match when randomization would make the groups unequal.


Matching by Equating Participants

o The most common method of matching is to equate participants on one or more variables to be controlled – called precision control method and matched by correlated criterion design.
o This method is able to detect small differences by ensuring that the participants in the various groups are equal on at least the paired variables.
o One requirement is that the variable which the individuals are matched, must be correlated significantly w/ the dependent variable. >.5 or .6


Disadvantages of Matching by Equating Participants

- 1) Difficult to know which are the most important variable to match (Select those variables that have the lowest correlation with each other, but the most correlated with the DV)
- 2) The decrease in finding eligible matched participants as the # of variables used for matching increases.
- Matching effects generalizability – Researcher can only generalize results to other individuals having the same characteristic as the matched sample.


The Frequency Distribution Matching Method

- Matches groups of participants in terms of overall distribution of the selected variable or variables.
- # of participants lost, not as great as individual-individual matching.
- The disadvantage is that combination of variables may be mismatched in various groups when there is more than on variable.