Explore BrainMass

Scaling, examples of bad survey questions, reliable vs invalid


Scaling and Question Design

? Distinguish among nominal, ordinal, interval, and ratio scales.
? Define criteria for good measurement in research.
? Describe the characteristics of a well-constructed questionnaire.
? Discuss the differences between ranking, rating, and multiple-choice attitude scales.

1. What are some examples of bad survey questions need 3? Please explain why they are bad and then reword them. You should give at least three examples of questions that you consider to be "bad".

2. How can a research project be reliable but invalid? Please provide an example. What would you need to do to create validity?

Solution Preview

Distinguish among nominal, ordinal, interval, and ratio scales.

-Nominal data is categorical in nature. It is not numerical. For example, if you ask people what their favorite ice cream flavor is, they will answer with categories or words, not numbers.

-Ordinal data is ranked data. For example, when you are given 3 choices of ice cream flavors, you have to rank which one you like the 1st, 2nd and 3rd. There is no magnitude of difference between how much you like the different flavors

- Interval scale will rank the items and tell you the magnitude of difference between the items. For example, the Celsius scale of measuring temperature is a good example of interval scale. There is no "zero" on this scale. - For example, when the temperature is 0 it exists, it does not mean that there is no temperature. Another characteristic is that you can say that it is twice as hot when it is 10 degrees then 20 degrees.

- Ratio scale is similar to interval, but it does have a real zero point. For example, age would be ratio, since a person could be zero years old. Salary as well, since a person could have 0 income.

? Define criteria for good measurement in research.

- Firstly, our survey needs to construct such that it measures what it is supposed to measures. In other words, our measurements need to provide data that is reliable and valid. Reliability means that the results of the data that comes out of the study will be consistent each time it collected. If this is not the case, there might be problems with the way that the questions are worded, which will reduce the quality of the measurements.
The second element is that the test is valid - that the results can be applied to the general ...

Solution Summary

Reliable versus invalid scaling examples of bad survey questions are examined.