I. Q: How to determine data sources?
a. Two research questions: what kinds of information do I need to answer the research question? Should I use primary or secondary data sources-or both-to obtain that information?
b. Divide the research question into appropriate elements or factors of analysis and narrow the scope
c. The specific kinds of data and specific data sources must be identified for each element in the scope
d. Focus your attention on finding data directly related to those factors
II. Q: Why should a researcher consult secondary data sources before primary sources?
a. Secondary sources often provide information to help define the problem more clearly and to identify elements that should be investigated.
b. When the problem you are investigating is unique to your organization, primary data may be the only usable information
c. If adequate secondary data are available, you should use those data instead of spending the time and effort required to use primary sources effectively
III. Q: What are the three criteria for the data sources?
a. Validity: a measurement concept that refers to the extent to which differences revealed with a measuring tool represent true differences among the people or objects being measured
b. Reliability: measurement concept that refers to the consistency of results obtained with a measuring device
c. Practicality: refers to both the cost and the convenience of using a data source
IV. Q: What are the characteristics of a valid sample?
a. Accuracy: an accurate sample is free of bias; an accurate sample has no systematic variance
b. Precision: a precise sample has little sampling error
c. Sample size: sample size can affect the accuracy and precision of your inferences
V. Q: What are the four guides for determining sample size?
a. Where samples are to be broken into subsamples (males/females. Juniors/seniors, etc.), a minimum sample size of 30 for each category is necessary
b. If several variables are used in the research, the sample size should be several times as large as the number of variables measured in the study
c. For simple experimental research with tight experimental controls, successful research is possible with samples as small as 10 or 20
No comments:
Post a Comment