The paper provides analyses of different cases pertaining to research scaling techniques. First part of the analysis includes providing different measurement scales for specific research concepts. Second part centers on the development of measurement scales for specific issues/research questions. Third section includes a discussion of different questions, assessment whether they are 'good" questions or not. Recommended revisions of the questions are provided. The last section discussed non-probability sampling with application on a specific case cited.
Scaling Techniques
The following concepts could be measured using the different kinds of measurement scales:
Store customers
Nominal -- specific geographic location of store customers within the store's trade area (i.e., residential location)
Ordinal -- comparative ranking of the store vs. other stores (i.e., store a has better service than store B, etc.)
Interval -- shopping hours or hours of the day when customers visit or go to the store
Ratio -- household income group (actual or estimated amount but will be re-categorized as high, medium, or low)
Voter attitudes
Nominal -- political affiliation or preference (Democrat, Republican, Independent, Other)
Ordinal - preference for a political candidate (prefer Candidate a than Candidate B, etc.)
Interval - number of hours per day (or number of days in a week) spent on engaging in politics-related activities
Ratio -- voter age (can be aggregated later as age group)
Hardness of steel alloys
Nominal -- steel alloy brand, geographic location of manufacturer, steel quality
Ordinal -- steel alloy quality (brand a harder than brand B, etc.)
Interval -- number of hours (interval time) for the steel alloy product to physically show "stress" in a product test (or hardness test)
Ratio -- hardness or thickness grade as ascertained from the product test
Preference for a particular common stock
Nominal -- industries or markets of interest when it comes to choosing and purchasing common stock
Ordinal -- assessment of stock option offerings (stock options for industry/company a performing better than stock options for industry/company B)
Interval -- years of engagement/involvement in stock investments
Ratio -- amount invested on common stock every quarter (or every year)
Profitability of various divisions in the company
Nominal -- division in the company the respondent belongs to Ordinal -- comparative ranking of profitability across divisions (division a more profitable than division B, etc.)
Interval -- number of months in a year that division has reached its target profit quota
Ratio -- profits earned per quarter or annually per division
2. Customer Confidence study for Menu Foods
Constant sum scale
"Given a grocery budget of $50, how many percent or what proportion of that budget will you allocate to Menu Foods products?"
Likert scale
Respond to this statement: "Menu Foods products meet my standards on product safety and quality."
5- Strongly agree 4- Agree 3- Neither agree nor disagree 2- Disagree 1-Strongly disagree
Semantic differential scale
Please evaluate and rate Menu Foods products based on the following characteristics:
Safe
Unsafe
Stapel scale
Rating of Menu Foods brand in the following attributes/characteristics:
Safety
Quality
Trustworthy
Honest
+5
+5
+5
+5
+4
+4
+4
+4
+3
+3
+3
+3
+2
+2
+2
+2
+1
+1
+1
+1
-1
-1
-1
-1
-2
-2
-2
-2
-3
-3
-3
-3
-4
-4
-4
-4
-5
-5
-5
-5
Forced ranking scale
"How would you rate, in general, Menu Foods with respect to the safety of its products?"
5 -- Product safety of products highly meets the standards.
4 -- Product safety meets more than the basic standards, but not high enough to be commendable or exceptional.
3 -- Product safety meets the basic standards.
2 -- Product safety does not meet all the basic standards.
1 -- Product safety does not meet the basic standards.
3. Evaluation of the following survey questions:
Do you read the National Geographic regularly?
This question is incorrectly phrased if the objective is to get the respondent's frequency of reading National Geographic. "Regularly" is vague and does not provide the researcher an idea about how frequent (or "regular") the respondents' readership is of National Geographic. Thus, the correct phrasing would be, "How frequent do you read National Geographic?" with choices ranging from monthly, quarterly, or yearly.
What percentage of your time is spent asking for information from others in your organization?
Since the questions asks for the percentage or proportion of the respondent's "time," the researcher must specifically state the particular "time" s/he is referring to. The reference time could be during the daytime, during the eight-hour working hours (from 8am to 5pm or 9am to 6pm). Whatever time duration is arrived at, the researcher should at least determine (preferably from the respondent) the time duration that the respondent will use as a reference, so it is clearer whether the proportion or percentage given is more or less than the eight-hour period provided as an example earlier.
When did you first start chewing gum?
While this is a simple and good enough question to ask about the respondent's first usage and consumption of a chewing gum, there is a possibility that s/he might not recall the first time s/he chewed a gum. This would leave the respondent unsure that it would take time for him/her to answer the question, and could even lead to an unanswered question in the survey. To account for this uncertainty, the question could be phrased, "Based on what you can recall (or based on your recollection), when did you first start chewing gum?" While the answer could not be accurately true, the response provided would at least be considered as reported/recalled by the respondent.
How much discretionary buying power do you have each year?
The term "discretionary buying power" is too complex for the respondent to understand. At best, the researcher could ask the respondent if s/he could provide the percentages or proportion of expenses s/he incur every year, and identify the products or services these percentages/proportions are spent on every year.
Why did you decide to attend Big State University?
While this question is acceptable to ask the respondent, the researcher must also indicate within the question that s/he knows there could be more than one reason why the respondent decided to attend Big State University. The question can be rephrased as follows: "What are the reasons that made you decide to attend Big State University? What else? Anything else?" Probing questions are added to ensure that all possible answers are exhausted from the respondent.
Do you think the president is doing a good job now?
This question can be answered by a simple "yes" or "no." If the researcher is looking for more depth in the respondent's answer than a simple yes or no, then s/he could rephrase the question to, "Do you think the president is doing a good job now? Why? Why else? Anything else?" Probing questions are added to ensure that all possible answers are exhausted from the respondent.
4. Since there is no prior data that could give an indication of concertgoers' profiles, the best that the researcher could do using non-probability sampling is to provide a profile of concertgoers surveyed, noting that the profiles would not be representative of the concergoer/viewer population that has attended the concerts/events.
You’re 83% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.