Question

2. A Multi-Item (Summated Scale) Consist of a number of closely related items (questions or statements) whose responses are combined into a composite score to measure a construct. Explain the benefits of measuring a construct with a multi-item scale as opposed to a single item scale in the context of an interview or questionnaire

3. Explain what is wrong with the following interview question and propose a solution to the problem you identify (i.e., re-write the question). (4 points)

a. Considering the high rate of childhood obesity and the fact that many Canadians eat fast food, don’t you think that fast food advertising in Canada should be banned?

b. Given the increased incidence of morbidity and mortality associated with gun ownership would you support increasing regulation of gun ownership and increased security in gun storage?

4. We are developing an online test to give to job applicants to measure their suitability for a sales position and one of key attributes for a good salesperson is ‘proactiveness’. To measure this, we have decided to use 24 different questions. Explain how you would use split-half reliability to ensure the questions were a reliable measure?

5. We are developing an online to give to job applications to measure their suitability for a sales position and one of the key attributes for a good salesperson is emotional intelligence. We plan to use Trait Emotional Intelligence Questionnaire (TEIQue) to measure emotional intelligence. Explain how and why you would use convergent validity to validity the TEIQue before using it?

6. An IT company operating in Atlantic Canada with 4 locations (Charlottetown, Halifax, Moncton, St. Johns) has seen its profits drop in recent years due to decreasing sales numbers. The company’s CEO believes that this is due to employees’ poor sales skills and has hired you to help address this problem. You perform design an experiment where half the sales staff at each location receive three weeks of sales training in the straight-line selling system. After training is completed, you track sales in both stores over a period of six months to assess the impact of sales skills training on store performance. Answer the following:

a. State an appropriate null hypothesis. (1 mark)

b. State an appropriate alternative hypothesis. (1 mark)

c. Describe a Type I error in the context of this study. (1 mark)

d. Describe a Type II error in the context of this study. (1 mark)

e. What is an important step that should have been completed before the experiment was designed? (1 mark)

f. Identify the dependent, independent and mediator variables

7. Comment on the descriptive statistics (i.e., means, standard deviations) and internal consistency reliabilities in the table below adapted from Erdogan and Bauer (2009) and summarize the correlation results, taking into account the strength, direction, and statistical significance of the bivariate relationships. (6 points)

Answer 1

1)benefits of using a multi-item scale over the single-item scale to measure the construct of interviews or questionnaires.

• internal consistency can be measured.
• not vulnerable to random measurement error.
• meaning and interpret quite accurately.

2)given interview questions are not straight and also lengthy. to remove the problem we make these question short and to the point included needed information.

a) Due to childhood obesity don’t you think that fast food advertising in Canada should be banned?

b) Due to the increase in morbidity and mortality associated with gun ownership, is it valid to increase gun storage security and regulation of gun ownership?

3)test- for a sales position and we check reliability on a key attribute of salesperson i.e proactiveness and for doing this we use split-half reliability test

• divide test in two halves by odd-even reliability.
• find the correlation of score finds in both parts by using person r formula.
• now use spearman brown formula to check reliability again.

4) why to use construct validity before TEIQue?

Trait Emotional Intelligence Questionnaire (TEIQue) emotional perception check by this.we use convergent validity before applying TEIQue because it tells a degree by which two measures of interview construct that theoretically related how much, in reality, these measures related. for example, we measure convergent validity on self-esteem now measure of similar constructs such as confidence, self-appraisal is related to self-esteem or not if yes then to what degree.

we use contract validity before applying TEIQue because we measure emotional perception and before reaching to any conclusion we must know the minute difference between similarly related words, that's why we use it before TEIQue.