Class Notes (786,419)
Canada (482,179)
Psychology (5,812)
Lecture 19

Lecture 19 - Quasi Experiments 2.docx
Lecture 19 - Quasi Experiments 2.docx

6 Pages
80 Views
Unlock Document

School
Western University
Department
Psychology
Course
Psychology 2800E
Professor
Doug Hazlewood
Semester
Fall

Description
Quasi Experiments 2: Studying the Effects of “Age” in Development Designs Prologue: Age as an IV A. Age is a “quasi” independent variable: - A participant variable that can’t be manipulated with random assignment B. Unlike gender, age changes over time: - “Males” don’t become “females”, - BUT, young people do become older. So, - Age can be a between-group factor (compare different “age groups” at same point in time) - Age can be a within-group factor (compare same group as they “get older” over time) C. Age introduces a unique set of confounds - Is the effect due to: o Age? (number of years person has lived) o Year of birth? (if different age groups are studied at the same time) o Time of testing? (if same group is studied at different times) Part 1: Cross­Sectional Design - Study different “age groups” at the same time o A between-group design Time of testing:2004 Group 1 Age = 20 Group 2 Age = 40 Group 3 Age = 60 - Major Problem: Confounds Age and Year of Birth o Called a “cohort effect”: people born in different years have different ages AND are in different generations (that have different “histories”) o Differences could be a result of chronological age or the fact that they come from different generations with different past histories - Example (time of testing = 2004) “Year of Birth” Cohort Age History 1984 20 TV, computers, email, internet, clel phones 1964 40 TV, no computers… 1944 60 No TV, only radio - Differences between groups could be due to different ages (an “age” effect); OR - Different “histories” (a cohort effect) - Solution to problem of cohort confounds? o Eliminate cohorts. Basic idea underlying... Part 2: Longitudinal Design - Study a single “cohort” over time o A within-group design Time of Testing 2004 2024 2044 Cohort (1984) Age: 20 40 60 - Problems: 1. Take a long time to complete (e.g., 40 years!) 2. Participant mortality: some participants die; o Some participants may drop out, or may lose track of them over time  Can’t generalize to the population 3. Introduces a new confound: o A Time of Testing effect (secular trend): Other things in the world might changebetween the times of testing (2004 to 2024 to 2044)  So “age” is confounded with “time of testing” o E.g., found that over time, people’s reflexes get slower. Is this due to age? OR environmental pollution got worse over time, making everyone sicker and slower (including 20 year olds in 2024 and 2044 who weren’t studied)  Not an age effect, reflexes decline may decline with environmental pollution which occurred over the time of testing - Q: How can we separate “age” effects form cohort and time of testing confounds? - A: Combine cross sectional and longitudinal designs in a single study Part 3: Cross­Sequential Design A. An example (W&M, Table 13.4, p. 330) Time of Testing Cohort 2000 2010 2020 1960 40 50 60 1970 30 40 50 1980 20 30 40 1990 10 20 30 - “Columns” = cross sectional (confounds age and “cohort”) - “Rows” = longitudinal (confounds age and time of testing) - Also, ”Diagonals” highlight a Time Lag Design o Different cohorts that are the same age, but tested at different times - Note: time lag designs don’t tell us about “age” effects (Age is held constant) o Time lag effects confounds “cohorts” and “time of testing” o BUT, they provide useful information:  The confounds are the same as the confounds in each of the other two design
More Less

Related notes for Psychology 2800E

Log In


OR

Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.

Submit