Market Research- Topic 6

What does marketing research measure?

Consumers’ perceptions, attitudes, preferences, characteristics — not the consumers themselves.

What is measurement?

Assigning numbers to characteristics of objects according to prespecified rules.

What is scaling?

Creating a continuum on which measured objects are located.

What is a nominal scale?

A labeling scheme where numbers act only as tags; no quantity implied.

What statistics can be used with nominal data?

Counting, percentages, mode.

What is an ordinal scale?

A ranking scale showing relative order, not magnitude of differences.

What statistics can be used with ordinal data?

Counting + centile measures (median, quartiles).

What is an interval scale?

A scale where equal numerical distances represent equal differences in the characteristic.

What statistics can be used with interval data?

All nominal + ordinal stats, plus mean, standard deviation.

What is a ratio scale?

The highest scale; includes classification, ranking, equal intervals, and an absolute zero.

What statistics can be used with ratio data?

All statistical techniques.

What is paired comparison scaling?

Respondents choose one option from each pair of objects.

When is paired comparison commonly used?

When comparing physical products.

What is rank order scaling?

Respondents rank several objects simultaneously based on a criterion.

What type of data does rank order produce?

Ordinal data.

What is constant sum scaling?

Respondents allocate a fixed number of points (e.g., 100) across attributes based on importance.

What does allocating twice as many points imply?

The attribute is twice as important.

What is a continuous rating scale?

Respondents mark a position on a line between two extremes.

How are CRS scores typically treated?

As interval data.

What characterizes itemised rating scales?

Categories with numbers or brief descriptions.

What are the two main itemised scales?

Likert and Semantic Differential.

What does a Likert scale measure?

Degree of agreement/disagreement with statements.

How many points does a typical Likert scale have?

Five (Strongly disagree → Strongly agree).

What is a semantic differential scale?

A 7‑point scale with bipolar adjective pairs (e.g., Powerful ↔ Weak).

How is semantic differential data analyzed?

Through profile analysis.

What is the recommended number of scale categories?

Between 5 and 9.

What is a balanced scale?

Equal favorable and unfavorable categories.

When use an unbalanced scale?

When the distribution is skewed.

What is the effect of using an odd number of categories?

Provides a neutral midpoint.

What is forced vs. nonforced choice?

Forced = no “no opinion” option.

What aspects of scale description can vary?

Verbal, numerical, pictorial labels.

What physical forms can scales take?

Vertical, horizontal, boxes, lines, continua.

What is a construct?

The characteristic being measured by a multi‑item scale.

What are the three criteria for evaluating a scale?

Reliability, validity, generalizability.

What is reliability?

Consistency of results across repeated measures.What is validity?

What is validity?

The extent to which observed scores reflect true differences, not error.

What is generalisability?

Ability to extend findings beyond the observed sample.

What type of question is the music‑genre exercise?

Structured.

What primary scale of measurement does it use?

Interval.

What scaling technique does it use?

Non‑comparative, itemised, Likert.