Methodology: Research Designs and Challenges

Question 1

What are our sources of information?

Answer 1

• Official records: CPS records. direct ratings from social workers, coding of records
• Self-report: Questionnaire or interview, parent report, or the self
• Observation: Live or video-recorded assessment, in the home, in the lab

Question 2

What is the best approach to measurement?

Answer 2

Multi-method, multi-source

Collect from various sources, pay attention to the measures being used and to the limits of the measurement, and pay attention to the way things are defined

Question 3

What are three research challenges?

Answer 3

1. Measurement: Difficult to quantify maltreatment in a standardized way, discrepancies across source of definitions
2. Testing conditions: Cannot experimentally manipulate abuse or neglect, which interferes with drawing causal inferences
3. Confounds: Difficult to parse maltreatment from co-occurring risk factors… and changes to the environment

Question 4

What are the advantages and disadvantages of CPS records? (Measurement)

Answer 4

Advantages: Often contains detailed info about incidents and can code for subtype and severity

Disadvantages: May miss cases that aren’t reported/investigated
Reliance of CPS labels can be based and uninformative

Question 5

What are the advantages and disadvantages of self-report? (Measurement)

Answer 5

Advantages: May obtain information that is not observable, or not commonly recorded in records and it easier to obtain

Disadvantages: Under-reporting due to fear of negative consequences and retrospective reports subject to recall bias

Question 6

What are the advantages and disadvantages of observation? (Measurement)

Answer 6

Advantages: More objective than self-reports, can set a sense of caregiving behaviors, can link to intervention

Disadvantages: Short in duration, so unlikely to see low base rate indicidents
Changes in behavior due to being observed

Question 7

What are the challenges of testing causation?

Answer 7

Correlation does not equal causation

We would need experimental manipulation to show that one thing causes another so we can do this by random assignment

In this case it would be unethical to set a random group of children to experience maltreatment

Question 8

What is a cross-foster design?

Answer 8

Example: You have a maltreating mother and offspring and a control mother and offspring and switch the two babies so that the maltreating offspring is with the control mother

You are separating nature vs nurture

The baby raised in a maltreating environment is likely to do the same

Question 9

What are confounds?

Answer 9

Confonds are third variables that could be associated with maltreatment and associated with the outcome of interest

SO are the negative effects of maltreatment actually due to maltreatment or are they just due to poverty