Scale construction and evaluation

Question 1

What is the goal of assessments in developmental research?

Answer 1

Identify risk factors - Predictors
Identify individuals on the basis of those risk factors

Question 2

What scale did Landa et al (2012) use to assess the development of children who’s siblings had Autism?

Answer 2

Mullen Scales of Early Learning

Question 3

What was the age range that Landa et al (2012) investigated? What was the design of their study?

Answer 3

6-36 months
Developmental trajectory

Question 4

What are the features of an assessment?

Answer 4

Standardisation
Concurrent and Predictive Validity
Flexibility
Range of content

Question 5

What is the key difference between assessments and questionnaires/checklists/inventories when it comes to evaluation and construction?

Answer 5

Focus more on who is reporting what when looking at questionnaires/checklists/inventories

Question 6

What are the benefits of parent reports?

Answer 6

Might not have the staffing or budget
Parents are with their children every day in comparison to a researcher who is with them for a short period
Valuable data - quantity and quality observations

Question 7

What are the drawbacks of parent reports

Answer 7

Can be unreliable observers
Systematic errors due to biases the parents have on their children
Exclusive use can lead to shared method variance

Question 8

What is shared method variance?

Answer 8

When using exclusively one method of report (whether that be parent or teacher or self) then you have to consider that a significant amount of the variance can be attributed to the reporter itself and not to the measure

Question 9

What example of SMV does Seifer (2006) use?

Answer 9

Child temperament and behaviour problems
Parents are reporters for both domains of child behaviour so the relationship may be due to using the parent for both domains and not the child

Question 10

How did Murphy et al (1999) address the issue of shared method variance in the study of children’s emotionality and regulation? What did they find?

Answer 10

Used both parent and teacher reports
Gave correlations between the two
There were correlations between them but parents were generally more consistent

Question 11

What 3 things caused variance in correspondence between cross-informants?

Answer 11

Context
Child Age
Observability of issues
i.e those that saw the child in similar contexts/ages had similar results

Question 12

What two questions should precede scale construction?

Answer 12

What are the existing measures?
Is a new measure really needed?

Question 13

What are the fundamental issues for scale constructions?

Answer 13

Measuring what you’re supposed to be? - Validity
Do relations b/ween variables/questions reflect some other causal factor?
Have you considered limitations to your conclusions?

Question 14

What are some threats to validty

Answer 14

Maturation, history, selection bias, testing effects, statistical regression, low reliability, mono-operation bias, mono-method bias

Question 15

What are the 8 steps of scale construction

Answer 15

1. Determine clear what you want to study
2. Generate an item pool
3. Determine format for measurement
4. Review item pool
5. Consider inclusion of validation items
6. Adminitster items to pilot sample
7. Eval items
8. Produce final scale

Question 16

What was the Baby Care Questionnaire interested in measuring?

Answer 16

Beliefs parents have that guide the care they give to their infants
Attunement and structure (cog and communicative development)

Question 17

What information drove Gattis et al to create the Baby Care Questionnaire

Answer 17

Infants born pre-term are considered at risk but early social interactions can be proven to be protective or detrimental. An association has been found between pre-term babies and cog/comm issues and it is a lot to do with parent beliefs

Question 18

What should be considered when generating an item pool

Answer 18

The item pool is large enough to allow for dropping items later
Clearly defined target construct to guide item selection
Items should be brief and readable
Include both positively and negatively worded items to evaluate bias

Question 19

What should be considered if you are thinking of dropping items?

Answer 19

See which questions gave strong expressions (really agree/really disagree)
Vivid items are good
Good negative items
What sources people use for parenting advice

Question 20

What is the issue with using likert scales?

Answer 20

It is ordinal data which can limit what you do with it statistically so it better to use interval data

Question 21

What did Chambers and Johnston (2002) find about children using Likert scales?

Answer 21

Should be mindful as the younger they are, the more extreme responses you get, especially with subjective times

Question 22

What procedures can you do to help with reviewing the item pool

Answer 22

Cognitive interview
Focus groups
Factor analysis

Question 23

What is factor analysis

Answer 23

Analytic tool that can help tell us about the properties of a scale (like stat analysis for each item);
determine how many factors or variables underlie a set of items
condense information
See how items cluster together

Question 24

What might be a better alternative to a likert scale and why?

Answer 24

Visual analog - offers greater accuracy of measurement for continuous constructs

Question 25

What is measurement equivalence

Answer 25

Ensuring the underlying constructs/factors in each group such as age, sociocultural