Assignment 6

Measurement Scales:

Thus far, you’ve been using what I call summary variables. These summary variables are actually means computed from the individual responses to multiple-item measurement scales. In this assignment, you’ll be working with the variables market orientation (MO) and operational performance (OP). Three hundred and twenty-eight managers working for U.S. manufacturing firms responded to a 10-item market orientation scale and an 8-item operational performance scale. Market orientation is a measure of customer focus and the organization’s ability to generate and disseminate information related to changes in customer demand. Operational performance is a form of organizational performance that focuses primarily on an organization’s efficiency. The actual measurement scales are provided below. Each of the managers was asked to read each item and indicate the degree to which they disagree or agree with the item. If the manager strongly disagrees, he or she is likely to put a 1. If the manager strongly agrees, he or she is likely to put a 7. If their agreement or disagreement is not so strong, they will respond somewhere in between 1 and 7.

Market Orientation Measurement Scale

Operational Performance Measurement Scale

When using measurement scales to collect data, you must use scales that are both reliable and valid. Validity means that the scales measure what they are supposed to measure like market orientation and operational performance. Reliability means that the scales measure consistently across similar samples. Each time measurement scales are used, they must be assessed reliability and validity.

Validity is assessed first. There are many types of validity. We’ll be assessing content, discriminant and convergent validity. Content (or face) validity means that the scale items look like they are measuring what they are supposed to measure. There is no statistical way to determine content validity. Just read the items and see if they are all related to what is being measured. Do the items in the market orientation scale seem to relate to market orientation? Do the items in the operational performance scale seem to relate to operational performance? If they do, conclude that the scales exhibit sufficient content validity. Discriminant validity means that the items in one measurement scale are measuring something distinctly different from the items in another measurement scale. We assess discriminant validity using a statistical technique called factor analysis. Convergent validity means that all the items in a measurement scale are measuring the same thing. We assess convergent validity using factor analysis as well. So, factor analysis is the key to determining discriminant and convergent validity. Once validity has been established, reliability can be assessed. The simplest way to measure reliability is to have SPSS compute Cronbach’s Alpha values.

It is sometimes necessary to remove scale items if they are found to hinder validity or reliability. The results of the exploratory factor analysis and the Cronbach’s Alpha computation will tell you which, if any, items should be removed. Finally, you average the remaining items to get the summary variable values.

The new part of this assignment requires assessing the market orientation and operational performance scales for validity and reliability and computing the summary variables. You’ll also be doing descriptive statistics, correlation, regression, and path analysis as you’ve done in previous assignments.

Complete the following steps:

1. Access the Assignment_6_Scale_Assessment_Data.sav file. Note that there are 10 items in the MO scale and 8 items in the OP scale.
2. Assess the degree of content validity by reading the items in each scale and noting that they relate only to the variables (MO or OP) that they are intended to measure. The 10 items in the MO scale all relate only to market orientation and the 8 items in the OP scale relate only to operational performance. So, conclude that the scales exhibit sufficient content validity. This is really all you can do to assess content validity; there is no way to statistically assess content validity. State your conclusion related to content validity below.
3. Assess for discriminant and convergent validity by running a confirmatory factor analysis using the following command sequence [analyze, dimension reduction, factor, move all of the items (both the MO and OP items) to the variables block, click on extraction, select "maximum likelihood" as the method, select "number of factors" as the extract type, enter 2 as the number of factors, continue, rotation, select "varimax", continue, okay] . Copy the r otated factor matrix (this is the next to last table in the output) to a Word document. Note that the highest factor loadings for the items in the MO scale are in the factor 1 column, and that the highest factor loadings for the items in the OP scale are in the factor 2 column. When all of the items separate nicely into factors according to scale (MO or OP), the measurement scales are considered to exhibit sufficient discriminant validity and convergent validity. In other words, the items in each of the scales converge into a single factor and are discriminated from the items in the other scales. So, the scales exhibit sufficient convergent and discriminant validity. State your conclusion related to convergent and discriminant validity below.
4. Assess the reliability of each of the scales by following the command sequence [analyze, scale, reliability analysis, block off the items for the MO scale only and move them to the items block, okay]. Follow the sequence again for the OP items. Notice that you’ll have to clear the items block of the MO items before moving the OP items over. Alpha values less than .70 indicate an unacceptable level of reliability; values greater than or equal to .70 indicate an acceptable level of reliability. Copy the two "Reliability Statistics" tables to the Word document. State your conclusion related to the relia bility of the two measurement scales.
5. Compute the summary values for the MO variable by following this command sequence [transform, compute variable, type MO in the target variable block, type MEAN(MO1,MO2,MO3,MO4,MO5,MO6,MO7,MO8,MO9,MO10) in the numeric expression block, okay]. Note that a new MO column has been added to the data set.
6. Compute the summary values for the OP variable by following this command sequence [transform, compute variable, type OP in the target variable block, type MEAN(OP1,OP2,OP3,OP4,OP5,OP6,OP7,OP8) in the numeric expression block, okay]. Note that a new OP column has been added to the data set.
7. Now, tie this to what you already know by generating a descriptive statistics table for the newly computed MO and OP variables . Note that you are now only using the summary values for MO and OP that you generated in step s 5 and 6 . Copy the table to your Word document and tell me whether the variables are normally distributed.
8. Generate a correlation matrix for the MO and OP variables. Note that you are now only using the summary values for MO and OP that you generated in step s 5 and 6 . Copy the table to you Word document and tell me type, strength, and significance level for the correlation coefficient for the relationship between MO and OP.
9. Generate the regression analysis for MO as the independent variable and OP as the dependent variable. Note that you are now only using the summary values for MO and OP that you generated in step s 5 and 6 . Copy the Model Summary and Coefficients table s to you r Word document . I nterpret the R ² value and the beta value.
10. Complete the path model by inserting the beta with significance level and the R ² value.
    

    Beta =

    
    

    R ² =

    OP

    MO

11. Based on the results, answer the following question: "Will a market orientation improve operational performance?"

Price: $25.93

Solution: The downloadable solution consists of 10 pages, 1593 words and 7 charts.
Deliverable: Word Document