Others

Question 1

how to determine number of groups in counterbalancing?

Answer 1

if its two conditions: (2X1)=2 groups if its three conditions: (3X2X1)=6 groups if its four conditions: (4X3X2X1)=24 groups

Question 2

latin squares

Answer 2

each treatment occupies one of the possible position on the block once and only once for example, for 5 conditions: A E C D B B A E C D C B A E C D C B A E E D C B A Or for 4 conditions: A B C D B C D A C D A B D A B C

Question 3

Cross sectional

Answer 3

between subjects-different samples are tested at different ages Disadvantage: they’re different people so the groups could just be different (cohort effect) because of some factor other than the one you’re studying

Question 4

cohort sequential

Answer 4

combination of within and between subjects-different samples are tested at multiple times as they age Start with group of 5 year olds, seven year old, and ten year olds Then follow them and retest when they are seven, nine, and twelve

Question 5

between subjects-need a control?

Answer 5

no control required because you can have treatment one and treatment two, so you don’t need a control as long as there is fair random assignment

Question 6

within and matched-control needed?

Answer 6

no control condition necessary individuals act as their own control in within, and their matched pair acts as their control in matched pair designs

Question 7

To say that A causes B, three conditions must exist (Mill, 1986):

Answer 7

• A precedes B

The cause must precede the effect temporally (X must come before Y in time order): Events that occur in the past can influence future events, and it is impossible for events in the future to influence events in the past.

• A and B must covary (A and B must be correlated)

There must be evidence that there is a consistent relation between the two variables. A difference between groups is evidence that the IV is correlated with the DV

• A must be the most plausible cause for B with other potential causes ruled out

All other possible explanations for the relationship between X and Y should be ruled out OR explained.

Question 8

regression discontinuity

Answer 8

RD design is a pretest-posttest program-comparison group strategy.

participants are assigned to program or comparison groups solely on the basis of a cutoff score on a pre-program measure.

Question 9

Longitudinal Designs types

Answer 9

1. Longitudinal
2. Cross-sectional
3. Cohort-sequential

Question 10

Things to keep in mind when creating a survey

Answer 10

• Keeping it simple
• Avoid double-barreled questions
• Avoid loaded/leading questions
• Avoid negative wording
• Avoid acquiescence
• Avoid vague response options
• Beware of order effects (response options/questions)
• Keeping scales consistent, avoid range restriction effects
• Ask sensitive questions sensitively
• Use questions relevant to all
• Striking the right balance for number of questions/items
• Miscellaneous: Where to place key questions, grouping items logically, considering lay-out, appropriate language and vocabulary for audience

Question 11

naturalistic observation

Answer 11

Observing individuals’ behavior in their normal environments.

The goal of naturalistic observation is to be unobtrusive so that the researcher does not affect the observed individuals’ behavior.

Often used in early phases of a research enterprise to identify important variables, and to suggest hypotheses about relationships NOT to establish cause and effect (no manipulation).

Some potential problems

Question 12

order effects

Answer 12

Order of Options: people favor option presented last