CO 600 Assignment 8

Question 12.6

You are working on a consumer perception study of four brands of bicycles. You will need to develop measurement questions and scales to accomplish the tasks listed below. Be sure to explain which data levels (nominal, ordinal, interval, ratio) are appropriate and which quantitative techniques you will use.

A) Prepare an overall assessment of all the brands.

B) Provide a comparison of the brands for each of the following dimensions: (1) Styling, (2) Durability, (3) Gear quality, (4) Brand image

Response from student # 1

The overall assessments of the four brands of bicycles will be done with a ranking scale. The measurement questions would be specific to the sub categories of styling, durability, gear quality, and brand image.  The first thing to look at is the study itself, whether it be a mail survey, interview, or web based questionnaire.  Spanjaard, Young, and Freeman explain that, “This work suggests ways that quantitative verifications can be (more) meaningfully informed by well-designed multiple method research.” (Spanjaard, Young, and Freeman, p.220, 2014)  This lends itself to various methods of data capture.  For this survey, consumer interviews at a local bicycle shop will be conducted as well as a web based questionnaire; distributed to bicycle consumers through email contact.  A forced choice rating scale with be used to address the overall brand assessment.  Zavala recommends, “The literature seems to indicate that raters prefer to choose statements either most or least descriptive, but not both; and that their resistance is further reduced by using positive rather than negative items.” (Zavala, p.121, 1965)  This psychosocial assessment of consumers is still valid 50 years later.  The measurement questions/statements will be derived from the four sub categories of comparison. This will integrate the entire questionnaire process and provide a variation in approaches to gather a more complete assessment.  The properties of the data level will be ratio scales to encompass all four property features. The comparison questions will be a choice scale scenario. The options will also be balanced in choice to provide a more complete look at the data levels. The options will be Best, Better than Average, Average, Worse than Average, and Worst.  These options with the measurement questions/statements will provide a solid ranking of all four brands.  These answer options will also be beneficial in the sub category comparisons.  The options will be tabulated from all surveys to present a brand comparison profile.  All the questions will be conducted from a multidimensional approach to consider the different areas of the bicycles.  One may have the best styling but may rank the lowest in gear quality.  The survey cannot assume that these are an all or nothing proposition.  Each brand may have its own strengths.  It is the overall assessment ranking that will provide the conclusion on which brand has the most positive feedback in all areas of questioning.               

The way a question is asked and the answers available to answer the question are just as important as the question itself when it comes to business research.  There is a science well explored to create the architecture for a well-balanced and meaningful survey.  Proverbs 2:2 states, “Making your ear attentive to wisdom and inclining your heart to understanding.”  The Lord tells us when we ask questions we must have an attentive ear in order to hear the answer.  He also tells us we must have an understanding heart with which to apply the knowledge that the questions provide.  All business research professionals should live by this Proverb to present accurate and complete data analysis. 

References

Spanjaard, D., Young, L., & Freeman, L. (2014). Emotions in supermarket brand choice: A multi-method approach. Qualitative Market Research: An International Journal, 17(3), 209-224. doi:10.1108/QMR-10-2012-0049

Zavala, A. (1965). Development of the forced-choice rating scale technique. Psychological Bulletin,63(2), 117-124. doi:10.1037/h0021567

Question 9.9

Response from student # 2

            An experiment must be conducted on 300 subjects to record what the response rates are of respondents when they are compensated for responding to an interview in comparison to when they are not compensated.  There are many various methods that can be utilized as a standard basis for the research design in order to conduct this experiment.   For this discussion, the researcher should choose to conduct a completely randomized design. The researcher is seeking to learn whether response rates are affected when the interviewed subject is exposed to monetary compensation.

            Three hundred respondents are going to be randomly selected to be personally interviewed about a number of social problems.   One hundred subjects in the sample should be randomly designated to receive no compensation.  One hundred other subjects should be randomly designated to receive $1 in compensation and the last one hundred subjects should be assigned to receive $3 in compensation. This research design can be best represented in the following diagram:

R1, R2, and R3 refer to the three random groups of 100 respondents each.  C0 refers to those receiving no compensation, C1 refers to those receiving $1 in compensation, and C3 refers to those receiving $3 in compensation.  N1, N2, and N3 refer to the number of responses received in that specific group.  This is a very simple method of research and it is all that is necessary in the case that has been presented.

            Whenever there is some choice about assigning experimental units to treatments, the units should be randomly allocated so that there is no hidden bias involved in the allocation. This then ensures a  

            fair comparison between the treatments even when there are important differences between the units that the experimenter is unaware of, and therefore cannot explicitly allow for in the experimental

            design. (Manly, 2010, p. 384).

The division of these 300 subjects must be equally distributed or else the research design will be flawed.  If it is learned that the research is lacking randomization; then a more complex research design may be warranted.  

            For the purposes of the research, it must also be considered that receiving compensation for a response may have a direct impact on the response that the subject being interviewed is going to provide the researcher with.  “Routine practices such as paying people to participate should be expected to have some impact on both these reasons for participation and possibly also their subsequent relationship with the research” (McCambridge, Kypri & Elbourne, 2014, p. 251).  This is indicative that certain variables may cause other unintended variables to surface through the course of the research.  However, it is impossible predict whether this will occur during a preliminary assessment of the design.

            The Bible reminds us “let each of you look not only to his own interests, but also to the interests of others” (Philippians 2:4, English Standard Version).  The respondents should want to help the researchers accomplish their goal of learning new information.  The Bible also mentions: “in everything I did, I showed you that by this kind of hard work we must help the weak, remembering the words the Lord Jesus himself said: ‘It is more blessed to give than to receive’” (Acts 20:35, New International Version).  Respondents of these surveys should not be enticed by receiving compensation.  This makes people focused solely on the acquisition of money rather than helping others to fulfill their mission of conducting research.

References