baselineone groupposttestonlydesignquasiexperimentaldesigncohortlongitudinalmethodsequentialmethodcross-sectionmethodcohorteffectsnonequivalentcontrol groupdesignone grouppretestpost testdesigncontrolseriesdesignfree butonly afterI call it!regressiontowardthe meanreversaldesignsingle caseexperimentaldesignselectiondifferencesinstrumentdecaytestingeffectsmaturationeffectsnonequivalentcontrol grouppretestposttestdesignhistoryeffectsinterruptedtimeseriesdesignpropensityscorematchingmultiplebaselinedesignsbaselineone groupposttestonlydesignquasiexperimentaldesigncohortlongitudinalmethodsequentialmethodcross-sectionmethodcohorteffectsnonequivalentcontrol groupdesignone grouppretestpost testdesigncontrolseriesdesignfree butonly afterI call it!regressiontowardthe meanreversaldesignsingle caseexperimentaldesignselectiondifferencesinstrumentdecaytestingeffectsmaturationeffectsnonequivalentcontrol grouppretestposttestdesignhistoryeffectsinterruptedtimeseriesdesignpropensityscorematchingmultiplebaselinedesigns

Chapter 11 Bingo (quasi-experimental) - Call List

(Print) Use this randomly generated list as your call list when playing the game. There is no need to say the BINGO column name. Place some kind of mark (like an X, a checkmark, a dot, tally mark, etc) on each cell as you announce it, to keep track. You can also cut out each item, place them in a bag and pull words from the bag.


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
  1. baseline
  2. one group posttest only design
  3. quasi experimental design
  4. cohort
  5. longitudinal method
  6. sequential method
  7. cross-section method
  8. cohort effects
  9. nonequivalent control group design
  10. one group pretest post test design
  11. control series design
  12. free but only after I call it!
  13. regression toward the mean
  14. reversal design
  15. single case experimental design
  16. selection differences
  17. instrument decay
  18. testing effects
  19. maturation effects
  20. nonequivalent control group pretest posttest design
  21. history effects
  22. interrupted time series design
  23. propensity score matching
  24. multiple baseline designs