CALIFORNIA STATE UNIVERSITY LONG BEACH PPA 696--RESEARCH METHODS: BINGHAM & FELBINGER CH. 2

1. BIBLIOGRAPHIC INFORMATION

1. Author: H. McKay, L. Sinisterra, A. McKay, H. Gong, & P. Lloreda

1. Title: Improving Cognitive Ability in Chronically Deprived Children.

1. Source: Science, V.200, 1987, pp.270-278.

 

1. SUMMARY OF THE RESEARCH
1. PROBLEM STATEMENT:

Can deprived children's cognitive abilities be improved?

1. BACKGROUND:

To assess both treatment for malnutrition and treatment for education, and whether greater amounts of treatment will produce greater and more enduring cognitive ability gains.

1. HYPOTHESIS:

The longer the duration of treatment, the greater the gains in cognitive ability.

1. MEASUREMENT OF VARIABLES

1. Dependent variable: cognitive ability

1. Independent variable(s): nutrition treatments; educational treatments

1. Control variable(s): n/a

1. RESEARCH DESIGN:

Pre-test/Post-test control group design (true experimental design); children exposed to from zero to four periods of treatment; one control group; one high SES comparison group.

Group	T1	Treat-ment 1	T2	Treat-ment 2	T3	Treat-ment 3	T4	Treat-ment 4	T5
G-1	O1	X	O2	X	O3	X	O4	X	O5
G-2	O1		O2	X	O3	X	O4	X	O5
G-3	O1		O2		O3	X	O4	X	O5
G-4	O1		O2		O3	nutri	O4	X	O5
G-5	O1		O2		O3		O4	X	O5
Control	O1								O5
Comparison	O1								O5

 

1. SAMPLING:

Randomized sampling and random assignment to treatment options; population of chronically deprived children in a poor neighborhood in Cali, Colombia.

1. INSTRUMENTATION:

21 different age-appropriate instruments employed over the life of the study to measure cognitive skills.

1. DATA COLLECTION/ETHICS:

Data collected at five points over the 4-year study by cognitive skills instruments; possible ethical concern for children left untreated (control group).

1. DATA ANALYSIS:

t-tests used to determine whether differences between the average cognitive scores of the 4 treatment groups, one control group, and one comparison group were statistically significant.

1. CONCLUSIONS:

Treatment does result in cognitive ability gains; the earlier treatment is begun, the higher the gains; the longer treatment continues, the higher the gains.

1. CRITIQUE

1. Possible Threats to Internal Validity

1. History:

Controlled by presence of control group; but a fire shortened the treatment group for year four (180 days; 185 days; 190 days; and 172 days)

1. Maturation:

Controlled by presence of control group; however, cannot separate out the effects of age when treated from the effects of the duration of the treatment in this study.

1. Testing:

Treatment group children could have improved from repeated testing; control and comparison groups tested only twice.

1. Instrumentation:

Instruments changed over time, may not be strictly comparable; were instruments translated from English or originally in Spanish?

1. Regression Artifact:

Children were chosen on the basis of low scores, could have improved over time without treatment.

1. Selection bias:

Population was limited to children whose parents agreed to participate in the study; could be some volunteer bias.

1. Experimental Mortality:

53 children (18%) dropped out over 4 years of the study from treatment groups; 44 (38%) from the control group; and 8 (21%) from comparison group. Dropouts could have been different.

1. Design contamination:

Children were entered into the program by neighborhood, limiting probability of design contamination.

1. Possible Threats to External Validity

1. unique program features:

Testing would have to be included in future treatments.

1. experimental arrangements:

Natural setting reduced effects of artificiality.

1. other threats:

May not be transferable to countries with different types of diets, languages, cultural customs, etc.