Quiz 2

Question 1

What is an operation definition?

Answer 1

One that defines a thing in such a way that it can be measured

Question 2

If you want to examine the impact of amount of education on voting behaviors, which one is the independent variable and which one is the dependent variable?

Answer 2

Amount of education is the IV

Question 3

Describe what an extraneous variable is (could be mediating, moderating, or confounding)

Answer 3

One that is not the focus of your study but that does have relationships with both the independent and dependent variables

Question 4

Give an operational definiton for driving under the inluence

Answer 4

Driving with a blood alcohol content of .08 or greater

Question 5

Give an operation definition of “sleep deprived”

Answer 5

Many answers are possible. Any one answer is right ass long as it related to the construct of intetest, is rational, and CAN BE MEASURED

Question 6

What is the difference between systematic error and random error? which is worse?

Answer 6

Random measurement error occurs unpredictably. Systematic measurment error occurs the same way everyime for either all pariticpants completeing a measure for some sb-group of the participants.
Systematic error is worse because it consistently biases estimation

Question 7

Explain what reliability is

Answer 7

Given that what is beaing measure has not actually changed, the measure of that construct is clsoe to the same every time the construct is measured

Question 8

What are two types of reliability, and explain what they mean

Answer 8

1. Test-Retest: you test a thing two times (close together in time) and get similiar answers
2. Inter-rater reliabilityL two different raters using the same instrument rate a phenomenon similiarlily

3, Internal consistency h’s reliability: a chronbachs alpha statistic indicated all the items in an instrument relate to the same construct

Question 9

what is validity

Answer 9

confidence that you are truly measuring the ocnstruct that you intend to measure

Question 10

what is face validity

Answer 10

seems reasonable “on the face of it”

Question 11

what is content validity

Answer 11

experts agree that the instrument captures the content that is necessary to measure and does not include content that does not relate to the construct

Question 12

what is criteron related validity

Answer 12

alligns with an external measure (criterion) of the construct. Could include predictive validity, concurrent validity, or known groups validity

Question 13

what is construct validity

Answer 13

a way to actually measure content validity by using other measurements to measure the construced being tapped (convergent) or onstruments that measure closely related but seperate constructs (discriminant).

Question 14

What are three criteria necessary to infer causality

Answer 14

1. the cause precedes the effect ( IV is established before hte DV)
2. the IV and the DV are correlated
3. Nothing other an the IV could have causes the change in the DV

Question 15

What is internal validity

Answer 15

confidence that you can infer causality

Question 16

What is a threat to internal validity?

Answer 16

maturation, history, selection effect, regression to the mean, testing effect, instrumentation changes, ambiguous temporal order

Question 17

What are the characteristics of a true experimental desighn (randomized control trial)

Answer 17

1) random assignment to at least 2 groups
2) measurement of the construct of interest before and afer delivery of an intervention
3) the intervention is delivered to only of the two groups

Question 18

Explain why true experimenta ldesign allows us to rule out threats to internal alisity

Answer 18

the design allwos us to measure change that would have taken place in the absensce of the intervention as well as change that is associated with the intervention

Question 19

What is a quasi-experimental design? How is it different from experimental design?

Answer 19

Quasi-experimental desgn: Utilizes at least 2 groups, only one of which gets the desire intervention. However, the two groups are not randomly assigned. Other quasi-experimental designs include multiple measures/time series designs

Question 20

Why is quasi-experimental design not as strong as / does not allow us to infer causality in the same way that eperimental design does?

Answer 20

Because the twp groups were not randomized, they may not be closely comparable. The groups may be non comparable on both measure characteristics and unmeasure characteristics. If this is so, those chaaracteristcs (rather than the independent variable) might have influence the change in thedependent variable

Question 21

What can you do to strengthen your quasi-experimental design?

Answer 21

1) ensure that the two groups are comparable on as many measures as possible
2) add more pretest/post test measures

Question 22

What kind of design is this? What is its strength at inferring causality and why? OXO

Answer 22

Pre-experimental, pretest/post-test design. Not strong because it does not include a comparison or control group- therefore we cannout rule out factors, such as maturation and historty, which might have been responsible for the change in the DV

Question 23

What would you do to the following design to strneth the study you are carrying out?

Answer 23

1) add a control group, preferably or at lease a comparson group matched closely to the intervention group so you can see how similair people performed over time in absence of the intervention
2) ad a pre-test so you can determine how the intervention group was functioning before and after the intervention took palce

Question 24

What is wrong with the design of this question:

Do you think it is untrue that higher education noes not teach students critical thinking?

Answer 24

it includes negatives

Question 25

What is wrong with the design of this question? Please rate your agreement with the following statement: graduate study should focus on teaching students to solve problems rather than focusing on teaching them specific skills. 1 = disagree2 = somewhat agree3 = agree4 = strongly agree

Answer 25

The Likert scale is unbalanced and does not include the same range between each rating level

Question 26

What is social desirability bias? What is acquiescence bias

Answer 26

Social desirability- unwilling to answer in a way that is not considered socially acceptable Acqueince- answers in a way that the participant beleives aligns with what the experimenter wants

Question 27

What is the Hawthorne effect

Answer 27

The participant's awareness that he/she is being observed causes him/her to behave differently.

Question 28

How does the topic of measurement relate to qualitative research?

Answer 28

Constructs are not operationally defined. Formal scales or instruments are not used. However, concerns about validity (trustworthiness) and reliablity (consistency) are still relvant

Question 29

______ is the process through which we specify precisely what we will mean when we use partiular terms

Answer 29

conceptualiation

Question 30

What is conceptual order

Answer 30

conceptualization ==> nominal definiton --> operational definition --> measurements in the real world

Question 31

the extenet to which we combine attributes in fairly gross categories

Answer 31

range of variation

Question 32

Problems in operationally definin variables

Answer 32

- we may not know in advance what all the msot salient variables are - Limited understanding of the variales may keep up from anticipating the best way to operationally define those variables - Even the best operational definitions are necesarrily superficial

Question 33

What is measurement errir

Answer 33

Data does not accuarely portray the concept that we attmept to measure

Question 34

What is systematic errr

Answer 34

when the information we collecte consistently reflets a false picture (it BAISES the measure)

Question 35

Ways to avoid measurement errorq

Answer 35

- use unbuased wording - carefully train intervewers - use unobtrusive observations to minimize the social desirabilty bias - Understand how exisiting records are kept - Triangulation

Question 36

the more reliable, the less...

Answer 36

random erorr

Question 37

acceptable test retest reliablity?

Answer 37

.7 or above

Question 38

PRedictive validity

Answer 38

measure can predict a criterion that will occur in the future

Question 39

what is concurrent validity

Answer 39

measure that correspons to a criterion that is known concurrently

Question 40

what is facotrial validitity

Answer 40

refers to whether the number of constructs and the items that make up those constructs measure what the researcher intends

Question 41

How do reliabiity and Validity work with qualitative research

Answer 41

Qualitative researchers study and describe things from multiple perspectives and meanings Less emphasis on whether one particular measure is really measuring what it’s intended to measure

Question 42

How to evaluate validity in qualitivative research (4)

Answer 42

The interpretation of parts of the text should be consistent The interpretation should be complete, taking all of the evidence into account The interpretation should be the most compelling one given the evidence within the text The interpretation should make sense of the text and extend our understanding of it

Question 43

Do pilot studies ahave a lot of internal validity?

Answer 43

no- meant to just generate information about an intervention that has little research

Question 44

3 common pre experimental desighsn

Answer 44

one shot case study one group pretest postest design postest only design with non equivalent groups

Question 45

problem with X O

Answer 45

one shot case study fails to control for any threats to internal validity

Question 46

01 X 02

Answer 46

one gtoup pretest post test design established correlation and time order but does not account for the factors otehr than the iv that might cause the change in the DV

Question 47

X 0 | 0

Answer 47

postest only design with noequivalen grouo cannot inger than any difference between the two groups was caused by the inervention

Question 48

what does randomization control for

Answer 48

selection bias in experimental designs

Question 49

what is research reacitivty and what are some examples of it

Answer 49

``` Refers to changes in outcome data that are caused by researchers or research procedures rather than the independent variable Measurement Bias Experimental Demand Characteristics Experimenter Expectancies Obtrusive Observation Novelty and Disruption Effects Placebo Effect ```

Question 50

What is attrition and how to avoid it

Answer 50

A threat to the validity of an experiment that occurs when participants drop out of an experiment before it is completed Strategies to minimize attrition: Reimbursement Avoid intervention or research procedures that disappoint participants Utilize tracking methods

Question 51

What is a nonequivalent compariosn groups design

Answer 51

- 2 existing groups tjhat appear to be similiar are identified or created - the dependent variable is assesed before and after an intervention is introduced to one of the groups - comparion group does not receive the intervention

Question 52

how to strengthen the validity of nonequivalent comparison groups design

Answer 52

- use multiple pretests | - use switching replication

Question 53

what is a simple time series design

Answer 53

a simple interrupted time series design attmepts to develpp causaul inferences based on a comparion of trends over mulriple measurements before and after the an intervention is introduced and requires no comparison group

Question 54

what is a multiple time series design

Answer 54

both an experimental group and a non equivalen comparison group are measure at multiple points in time before and after an intervention is introduced to the experimental group

Question 55

pitfalls in carrying out experiments and quasi experiments in social science agencies

Answer 55

Social work experiments tend to take place in agency settings where administrators are not researchers, may not understand the requisites of experimental and quasi-experimental designs, and may even resent and attempt to undermine the demands of the research design

Question 56

What is a case control study

Answer 56

compares groups of cases with contrasting outcomes and collects retrospective data that may expalin thedifferences in the outcome popular because of feasibility as data can be collected at one point in time inference and generalizabilty are limited

Question 57

4 common pitfalls commonly encountered when implementeing research in social service agencies

Answer 57

Fidelity of the intervention Contamination of the control condition Resistance to the case assignment protocol Client recruitment and retention