Class Notes (1,200,000)
CA (650,000)
UTSC (30,000)
Psychology (9,000)
PSYB01H3 (200)
Anna Nagy (100)

Oct 25

Course Code
Anna Nagy

This preview shows page 1. to view the full 5 pages of the document.
Oct.25th, 2010
Confound variable: An uncontrolled 3rd variable known to affect a relationship between two variables
Look at past studies to see if theres any confounding factors already noticed limitations of our study
GOAL: Find convergence in results among different studies
How? randomly assign participants to do experimental/ control design, equally distributing the
differences among participants
When? when the variables among participants cannot be held constant
Why? Sampling the whole population = impossible, so we sample 1 population most of the time
(limited sample size)
BUT, randomization may not be
Easy/ Enough because differences between subjects are too large
Evaluating Research:
Validity= truth and accurate representation of information
Construct Validity
Adequacy of the operational definition of variables
Are the results replicable?
Internal Validity
Ability to draw conclusions about causal relationships from our data
*Experimental studies have higher chance b/c were controlling variable, reducing possibilities of
Two general threats to internal validity:
1. when control groups are absent (non- experimental studies) b/c you have less power to
control study and come to conclusion based on collected date
2. when comparisons are made between nonequivalent groups (Quasi-experimental: you
take subject groups as they are e.g: where victims live?) chances that differences between
the subjects maybe large enough to affect results
You're Reading a Preview

Unlock to view full version

Only page 1 are available for preview. Some parts have been intentionally blurred.

External Validity = Extent to which the results can be generalized to other populations and settings
- Consider cohort effects (a group of ppl who were born in a specific time therefore experiencing )
the same life experience affecting them profoundly) /time of day effects
- Despite their artificial setting, lab experiments often have great value for understanding human
Reliability: Consistency or Stability of a Measure of Behavior after re-testing
Less Error
Calculated as Pearson r
- Correlation coefficients called regression coefficients (reliability coefficient) used for ratio and
interval scales
Coefficients range from 0.00 to - 1.00 (variables covary in opposite directions) and 0.00 to +1.00
(variables vary in same direction)
Testing for reliability:
1. Test-retest reliability:
-assessed by measuring the same individuals two points in time (subjects position in relative
distribution of scored should be similar to be reliable)
Vulnerable to artificially (participants start to figure out what the purpose of study it
Hence change behavior)
Vulnerable to maturation (changes due to natural maturation of participants)
2. Internal Consistency Reliability
You're Reading a Preview

Unlock to view full version