Research Methods in
Cognitive Psychology
A “Real Life” Example
1
Centre Daily Times – AP Article
• Preschool Prevents Crimes, group says
– Poll: Kindergarten teachers say pupils less
disruptive
• Group = Fight Crime: Invest in Kids
• Poll = Mason-Dixon Polling & Research –
National Kindergarten Teacher Survey,
2004
2
Fight Crime: Invest in Kids
• Founded in 1996, Fight Crime: Invest in Kids is a bipartisan,
nonprofit anti-crime organization led by more than 2,000 police
chiefs, sheriffs, prosecutors, victims of violence and leaders of
police officer associations.
• Our mission is to take a hard-nosed, skeptical look at the research
about what really works-and what doesn't work-to keep kids
from becoming criminals. We then put that information in the
hands of policy-makers and the public. Among the strategies
proven to be effective are early care and education programs for
preschoolers, good schools, after-school programs and
prevention of child abuse, as well as programs that nip
delinquency in the bud by getting troubled kids back on track.
• National operations and state offices in California, Illinois,
Maine, Michigan, New York, Ohio and Pennsylvania
• Fight Crime: Invest in Kids receives no funding from state or
federal governments.
• http://www.fightcrime.org 3
National Kindergarten Teacher
Survey
• Data collected July 9 – July 14, 2004 via
telephone
• 800 kindergarten teachers were interviewed
• Interviewees were randomly selected from a
commercially available list compiled from state
licensing records
• Margin for error - ± 3.5%
• Number of questions? Possibly 25
• http://www.fightcrime.org – PDF Poll
4
Interesting Findings
• Majority of teachers said fewer • Approximately 40% of the
than 20 of their pupils last year teachers responded that measures
were inadequately prepared for of learning were very important
kindergarten or essential to be ready for
• Majority of teachers said that 41% Kindergarten (measures
or more of their students had including counting to 20,
attended Pre-K programs; 42% said possessing problem solving
that 61% or more of their students skills, and knowing letters)
had attended Pre-K • Approximately 85% of the
• Approximately 99% of the teachers teachers responded that measures
said that the children attending Pre- of good behavior were very
K programs were equally or better important or essential to be ready
prepared to start school ready to for Kindergarten (measures
succeed. including not disruptive, gets
along with others, pays attention)
How much of your class time is spent dealing with child
misbehavior:
67% responded less than 25% of the time
34% responded more than 25% of the time 5
Other Findings
• How much more likely is a child • Children in your class last year
from quality pre-K program to who attended pre-K typically
possess the following will do better in school than
characteristics: those who did not attend pre-K
– Count to 20 – 92% – 86% agree
– Problem solving skills – 86% • Substantially more children
– Knows letters – 93% would succeed in school if all
families had access to quality
• Learning (90.3%)
pre-K programs
– Not disruptive – 78% – 93% agree
– Gets along with others – 88% • The time I devote to dealing
– Pays attention – 88% with disruptive behavior by
• Behavioral (84.67%) poorly-prepared children, and
helping them catch up,
negatively affects the progress
of well-prepared children
– 86% agree
6
Other Research to Consider
• High/Scope Perry Preschool – Ypsilanti, MI
• http://www.highscope.org/Research/PerryProject/perrymain.htm
– 123 African American 3 and 4 year olds; Experimental Group=58 children; Control Group=65 children
– Attended program for one or two years
– Attended the program 2.5 hours everyday from October to May; Teacher/student ratio was 1 to 5.7
– Home visits also occurred
• Data were collected on program participants and no-program participants every year between ages 3 and 11,
and then at ages 14-15, 19 and 27; Raw data not readily available
• General findings:
– By age 27, only one fifth as many preschool program group members as no-preschool program group
members had been arrested five or more times (7% vs. 35%), and only one third as many had ever been
arrested for drug dealing (7% vs. 25%).
– Almost a third again as many preschool program group members as no-preschool program group members
graduated from regular or adult high school or received General Education Development certification
(71% vs. 54%).
– Although the same percentage of preschool program males and no-preschool program males were married
(26%), the preschool program males had been married nearly twice as long as no-preschool program
males (averages of 6.2 years vs. 3.3 years). Five times as many preschool program females as no-program
females were married at the time of the age-27 interview (40% vs. 8%). Further, preschool program
females had only about two thirds as many out-of-wedlock births as no-preschool program females (57%
vs. 83%).
– At age 27, four times as many preschool program group members as no-preschool program group
members earned $2,000 or more per month (29% vs. 7%). Almost three times as many preschool program
group members as no-preschool program group members owned their own homes (36% vs. 13%); and
over twice as many owned second cars (30% vs. 13%). Only three fourths as many preschool program
group members as no-preschool program group members received welfare assistance or other social
services at some time as adults (59% vs. 80%).
7
Chicago Child-Parent Centers
• The Chicago Child-Parent Center (CPC) Program (n = 989 children) provides
comprehensive education, family, and health services and includes half-day
pre-school at ages 3-4 years, half- or full-day kindergarten, and school-age
services in linked elementary schools at ages 6-9 years.
• The comparison group (n = 550) consisted of children who participated in
alternative early childhood programs (full-day kindergarten): 374 in the pre-
school comparison group from five randomly selected schools plus two others
that provided full-day kindergarten and additional instructional resources and
176 who attended full-day kindergartens in six CPCs without pre-school
participation.
• The main outcome measures were rates of high school completion and school
drop-out by age 20 years, juvenile arrests for violent and non-violent offences,
and grade retention and special education placement by age 18 years.
Pre-K No Pre-K
High-School Completion 49.7% 38.5%
Years of completed education 10.6 10.2
Lower rates of juvenile arrest 16.9% 25.1%
Fewer violent arrests 9.0% 15.3%
Fewer school dropouts 46.7% 55.0% 8
Conclusions from Fight Crime:
Invest in Kids
• Recent long-term studies show that at-risk
children who attend quality early childhood
care and pre-kindergarten programs are far
less likely to become criminals than those
denied access to such programs.
• Poor quality early childhood care multiplies
the risk that children will grow up to
become criminals.
9
Interesting Quotes
Reeves confirmed that the poll accurately depicted her experience during
28 years as a kindergarten teacher: “I’ve seen 5-year-olds without pre-
k who couldn’t count to five. Some don’t know their full names. It’s
really sad. You can tell the children who are going to get in trouble
with the law. I’ve seen it happen.” She said high quality pre-
kindergarten gives children the chance to succeed in school. -
Knoxville, Tenn. kindergarten teacher Patricia Reeves; Fight Crime
News Release, August 11, 2004
Through its free preschool program, Georgia hopes to set poor children
on a course of academic achievement. . . Sanford Newman, president
of the advocacy group Fight Crime: Invest in Kids, understands that
taxpayers in Georgia and elsewhere will dismiss publicly funded
preschool for 3-year-olds as free day care. “They’re wrong,” he says.
“They should see it as an investment.” “When we know that quality
early childhood programs cut crime and increase school success, it’s
just common sense to make those investments,” he says. “Otherwise,
we’ll pay down the road in violent crime and disruptive students.” -
Atlanta Constitution 14 April 2001
10
Association to Promote Importance
of Learning over Behavior
• Perceived Problem = educators emphasize the importance of behavior over learning in students
as young as kindergarten age
• Mission = to stress, through research, to educators that methods of effective learning should be
their top priority in the classroom rather than spending a majority of their time on behavior
management.
• Supporting Research
• Fight Crime: Invest in Kids Kindergarten Poll
– How much of your class time is spent dealing with child misbehavior:
• 67% responded less than 25% of the time
• 34% responded more than 25% of the time
– Approximately 40% of the teachers responded that measures of learning were very
important or essential to be ready for Kindergarten (measures including counting to 20,
possessing problem solving skills, and knowing letters)
– Approximately 85% of the teachers responded that measures of good behavior were very
important or essential to be ready for Kindergarten (measures including not disruptive, gets
along with others, pays
• Bennett KJ, Brown KS, Boyle M, Racine Y, Offord D (2003). Does low reading achievement at
school entry cause conduct problems? Soc Sci Med. 2003 Jun 56(12), 2443-2448.
– The findings show that an eight point increase in reading scores (equivalent to an moderate
effect size of 0.5) would result in a 23 per cent decrease in the risk of conduct problems 30
months later, after controlling for gender, income and baseline conduct problem symptoms.
We conclude that reading problems may contribute to the early onset of conduct disorder.
11
Research Methods Issues
• What kind of study was the Kindergarten Poll?
– Descriptive
– Correlational
– Experimental
– Quasi-Experimental
• What are the major constructs to be defined?
• Do you think it is internally valid? Ecologically valid?
• Are there any biases present in this research?
• Is any of the research presented for this example
experimental?
12
Experimental Research
• Systematic manipulation and measurement of variables
• Observation of the effects of manipulated variables on
one another in controlled settings/conditions
• Allows for causal statements
• Regardless of the type of research method (even
experimental), no research study PROVES an outcome
– More research with similar outcomes allows us to make
statements that outcomes “are more likely to occur” (or less
likely) under the same conditions
• Primary components of experiments:
– Independent Variables (manipulated variables)
– Dependent Variables (measured variables) 13
Next Class
• Read Chapter 2 (if you haven’t already)
• Review material from previous courses in
Research Methods and Statistics
14