Question 1

A scale that is reliable and valid in one culture might not be reliable and valid in another culture.

Accepted Answer

A) True 
 B)False  True

Question 2

The questionnaire item "Which is not a serious problem in the U.S.today,anomie or the socialization of the young?" violates which of the following guidelines?

Accepted Answer

A)  Avoid double-barreled questions. 
B)  Avoid short items. 
C)  Avoid leading the respondent. 
D)  Avoid meaningful questions. 
E)  Avoid negative items. 
A)  Avoid double-barreled questions. 
B)  Avoid short items. 
C)  Avoid leading the respondent. 
D)  Avoid meaningful questions. 
E)  Avoid negative items. E

Question 3

Identify acceptable options for handling missing data in an index or scale.

Accepted Answer

When dealing with missing data in an index or scale, it is important to choose a method that is appropriate for the type of data, the amount of missing data, and the research context. Here are several acceptable options for handling missing data:

1. **Listwise Deletion (Complete Case Analysis)**: This method involves removing any cases (respondents or subjects) that have missing data on any variable included in the index or scale. It is simple to implement but can lead to a significant reduction in sample size and potential bias if the missing data are not completely random (i.e., missing completely at random, MCAR).

2. **Pairwise Deletion**: Unlike listwise deletion, pairwise deletion uses all available data by analyzing pairs of variables that have no missing values. This method maximizes the use of available data but can lead to different sample sizes for different analyses and potentially inconsistent results.

3. **Mean Substitution**: This technique involves replacing missing values with the mean of the available cases for that variable. While it is easy to implement and maintains the sample size, it can underestimate the variance and covariance of the data and may not be appropriate if the data do not meet the assumption of being MCAR.

4. **Multiple Imputation**: Multiple imputation is a more sophisticated technique that involves creating several different plausible imputed datasets and then combining the results from each. This method accounts for the uncertainty of the missing data and is less likely to introduce bias compared to simpler methods like mean substitution.

5. **Maximum Likelihood Estimation (MLE)**: MLE is a statistical approach that estimates model parameters in a way that would maximize the likelihood of observing the data that we have, including the missing values. It is often used in conjunction with structural equation modeling and can provide unbiased estimates under certain conditions.

6. **Last Observation Carried Forward (LOCF)**: In longitudinal studies, if a participant drops out, their last observed response can be carried forward to replace missing future responses. This method is easy to apply but can introduce bias if the assumption that the last observation remains stable over time is not met.

7. **Hot Deck Imputation**: This method involves replacing a missing value with an observed response from a similar individual or case. The "donor" case is typically selected based on a set of matching criteria, ensuring that the imputed value is reasonable given the data's structure.

8. **Model-Based Methods**: These include using regression models or other predictive models to estimate missing values based on the relationships observed in the data. The choice of model should be informed by the data structure and the underlying assumptions about the missing data mechanism.

9. **Sensitivity Analysis**: While not a method for handling missing data per se, conducting sensitivity analyses can help to assess how robust the results are to different assumptions about the missing data. This can involve comparing results across different methods of handling missing data.

When choosing a method, it is crucial to consider the assumptions each method makes about the nature of the missing data (MCAR, MAR - missing at random, or NMAR - not missing at random) and the potential impact on the results. The chosen method should be reported and justified in any research publication to ensure transparency and replicability.

Question 4

Which of the following statements is/are true about constructing qualitative measures?

Accepted Answer

A)  You should rely primarily on close-ended questions. 
B)  Highly structured self-administered questionnaires are more desirable than interview guides. 
C)  Interviews can range from completely unstructured, informal conversational interviews that use no measurement instruments to highly structured, standardized interviews in which interviewers must ask questions in the exact order and with the exact wording in which they are written out in advance. 
D)  None of these answers are true. 
A)  You should rely primarily on close-ended questions. 
B)  Highly structured self-administered questionnaires are more desirable than interview guides. 
C)  Interviews can range from completely unstructured, informal conversational interviews that use no measurement instruments to highly structured, standardized interviews in which interviewers must ask questions in the exact order and with the exact wording in which they are written out in advance. 
D)  None of these answers are true.

Question 5

The following items and Scoring scheme were taken from the Minnesota Survey of Opinions.The four items were part of an index designed to assess attitudes toward education.In the index
1 A man can learn more by working four years than by going to school,
$\begin{array} { l l l l l } \text { SA5 } 54 & \text { A } 4 & \text { D } 2 & \text { SD } 1 \end{array}$
2 The more education a man has the better he is able to enjoy life.
$\begin{array} { l l l l l } \text { SA } 1 & \text { A } 2 & \text { U } 3 & \text { D } 4 & \text { SD } 5 \end{array}$
3 Education helps a person to use his leisure time to better advantage.
$\begin{array} { l l l l l } \text { SA } 1 & \text { A } 2 & \text { U } 3 & \text { D } 4 & \text { SD } 5 \end{array}$
4 Education is of no help in getting a job today.
$\begin{array} { l l l l l } \text { SA } 5 & \text { A } 4& \text { U } 3 & \text { D } 2 & \text { SD }1 \end{array}$

Accepted Answer

A)  Likert scale. 
B)  Thurstone scale. 
C)  Guttman scale. 
D)  semantic differential scale. 
E)  Bogardus social distance scale. 
A)  Likert scale. 
B)  Thurstone scale. 
C)  Guttman scale. 
D)  semantic differential scale. 
E)  Bogardus social distance scale.

Question 6

Choose a topic not discussed in the book or in class and devise a scale using one of the following methods:a.Likert-type scaleb.Semantic differential scale
Be sure to discuss all the steps you would go through in constructing your scale.

Accepted Answer

The answer of Choose a topic not discussed in the...

Question 7

Which of the following statements is/are true about questionnaires?

Accepted Answer

A)  Researchers can avoid the need for pretesting if they are very careful in the way they word questions. 
B)  Because it is virtually impossible to avoid some mistakes in wording, questionnaires should be pretested in a dry run. 
C)  When pretesting, the pretest sample should be very large. 
D) When pretesting your questionnaire, it's better to ask people to read through it looking for errors rather than to complete the questionnaire. 
A)  Researchers can avoid the need for pretesting if they are very careful in the way they word questions. 
B)  Because it is virtually impossible to avoid some mistakes in wording, questionnaires should be pretested in a dry run. 
C)  When pretesting, the pretest sample should be very large. 
D) When pretesting your questionnaire, it's better to ask people to read through it looking for errors rather than to complete the questionnaire.

Question 8

Qualitative measures often rely on interviews that are usually open-ended with in-depth probes.

Accepted Answer

A) True 
 B)False

Question 9

The following answer categories about ethnicity are exhaustive: Caucasian ________ African American ________ Hispanic ________

Accepted Answer

A) True 
 B)False

Question 10

There are several formats for contingency questions.Give illustrations of three differenormats,using examples not presented in the text.How does the formaor contingency questions differ from the format used for matrix questions? Give an illustration of the latter to differentiate irom the former.

Accepted Answer

Contingency questions can be formatted i

Question 11

When the variable "marital status" is classified as never married,currently married,or currently married but separated,this variable has the important quality/ qualities of being

Accepted Answer

A)  mutually exclusive. 
B)  exhaustive. 
C)  interchangeable. 
D)  ratio scale. 
A)  mutually exclusive. 
B)  exhaustive. 
C)  interchangeable. 
D)  ratio scale.

Question 12

The phrasing used in closed-ended questions should allow respondents to attach their own meanings and interpretations to the question.

Accepted Answer

A) True 
 B)False

Question 13

The length of a question can influence how effectively the respondent will answer.

Accepted Answer

A) True 
 B)False

Question 14

In looking over the responses given to the items in question 8,you notice that only one of your 2000 respondents had missing data on any of the items.That respondent scored 3 on the first 3 items but left item 4 blank.You should:

Accepted Answer

A)  give the respondent a total score of 9. 
B)  assign the middle value or the respondent's average value to item 4, giving that respondents a total score of 12. 
C)  exclude that respondenrom your data analysis. 
D)  either exclude that respondenrom your analysis or assign the middle or average value to item 4 
A)  give the respondent a total score of 9. 
B)  assign the middle value or the respondent's average value to item 4, giving that respondents a total score of 12. 
C)  exclude that respondenrom your data analysis. 
D)  either exclude that respondenrom your analysis or assign the middle or average value to item 4

Question 15

The result of combining several indicators of a variable is a composite measurement of the variable.

Accepted Answer

A) True 
 B)False

Question 16

Asking respondents if they have any children before asking them about the ages of their children is an example of:

Accepted Answer

A)  a contingency question. 
B)  a double-barreled question. 
C)  a biased question. 
D)  a negative item. 
A)  a contingency question. 
B)  a double-barreled question. 
C)  a biased question. 
D)  a negative item.

Question 17

Identify measurement principles that always apply to quantitative and qualitative approaches to asking people questions.Then identify differences between the two approaches in the style of asking questions.

Accepted Answer

Measurement principles that always apply

Question 18

The questionnaire item "Did you file federal and state income tax reports last year?" with a response set of yes,no,can't remember,other,is an example of

Accepted Answer

A)  an open-ended question. 
B)  an ordinal variable. 
C)  a double-barreled question. 
D)  a negative item. 
A)  an open-ended question. 
B)  an ordinal variable. 
C)  a double-barreled question. 
D)  a negative item.

Question 19

In looking over the responses given to the items in question 8,you notice that a respondent checked "strongly agree" to all four items.As a methodologist,which of the following statements BEST describes your ideas about the respondent?

Accepted Answer

A)  The respondent is a scale type 12. 
B)  The respondent probably fell into a response set. 
C)  The respondent is probably unsure about his or her feelings toward education. 
D)  The respondent must strongly agree with all the items. 
E)  The respondent has a high opinion of education. 
A)  The respondent is a scale type 12. 
B)  The respondent probably fell into a response set. 
C)  The respondent is probably unsure about his or her feelings toward education. 
D)  The respondent must strongly agree with all the items. 
E)  The respondent has a high opinion of education.

Question 20

Always include at least one item in an index for which all respondents give the same answer.

Accepted Answer

A) True 
 B)False

Exam 10: Constructing Measurement Instruments

A scale that is reliable and valid in one culture might not be reliable and valid in another culture.

The questionnaire item "Which is not a serious problem in the U.S.today,anomie or the socialization of the young?" violates which of the following guidelines?

Identify acceptable options for handling missing data in an index or scale.

Which of the following statements is/are true about constructing qualitative measures?

Choose a topic not discussed in the book or in class and devise a scale using one of the following methods:a.Likert-type scaleb.Semantic differential scale Be sure to discuss all the steps you would go through in constructing your scale.

Which of the following statements is/are true about questionnaires?

Qualitative measures often rely on interviews that are usually open-ended with in-depth probes.

The following answer categories about ethnicity are exhaustive: Caucasian ________ African American ________ Hispanic ________

There are several formats for contingency questions.Give illustrations of three differenormats,using examples not presented in the text.How does the formaor contingency questions differ from the format used for matrix questions? Give an illustration of the latter to differentiate irom the former.

When the variable "marital status" is classified as never married,currently married,or currently married but separated,this variable has the important quality/ qualities of being

The phrasing used in closed-ended questions should allow respondents to attach their own meanings and interpretations to the question.

The length of a question can influence how effectively the respondent will answer.

In looking over the responses given to the items in question 8,you notice that only one of your 2000 respondents had missing data on any of the items.That respondent scored 3 on the first 3 items but left item 4 blank.You should:

The result of combining several indicators of a variable is a composite measurement of the variable.

Asking respondents if they have any children before asking them about the ages of their children is an example of:

Identify measurement principles that always apply to quantitative and qualitative approaches to asking people questions.Then identify differences between the two approaches in the style of asking questions.

The questionnaire item "Did you file federal and state income tax reports last year?" with a response set of yes,no,can't remember,other,is an example of

In looking over the responses given to the items in question 8,you notice that a respondent checked "strongly agree" to all four items.As a methodologist,which of the following statements BEST describes your ideas about the respondent?

Always include at least one item in an index for which all respondents give the same answer.

Why Study Research

Evidence-Based Practice

Philosophy and Theory in Social Work Research

Comparing Quantitative, Qualitative, and Mixed Methods of Inquiry

The Ethics and Politics of Social Work Research

Culturally Competent Research

Problem Formulation

Conceptualization in Quantitative and Qualitative Inquiry

Measurement

Causal Inference and Experimental Designs

Quasi-Experimental Designs

Single-Case Evaluation Designs

Program Evaluation

Sampling

Survey Research

Analyzing Existing Data: Quantitative and Qualitative Methods

Qualitative Research: General Principles

Qualitative Research: Specific Methods

Qualitative Data Analysis

Quantitative Data Analysis

Inferential Data Analysis

Inferential Data Analysis

Writing Research Proposals and Reports

Filters

The following answer categories about ethnicity are exhaustive: Caucasian African American Hispanic