Читать книгу Social Work Research Methods - Reginald O. York - Страница 86

Step 3: Collecting and Analyzing Data

Оглавление

You will collect data according to the relevant protocol. The term data refers to the discrete information that you have, such as the age of each client, the depression score for each client, and so forth. You must have data for each variable in your study. You cannot have a research question that contains age as a variable unless you collect data on age for each study subject.

Your data collection procedure may mean giving a questionnaire to a group of people at one point in time. It might entail the administration of a tool more than once to the same group of people. In this book, you will have exercises where you will collect data from a group of people (perhaps the members of your class) at one point in time. The instrument used will have items designed to measure each of the variables in your study. One of the variables may designate the group the respondent is in, so that two groups can be compared.

You will select a statistic for each of your research questions. Some of these statistics will be descriptive in nature, such as a frequency or a mean. Some will be explanatory in nature because the statistic helps you determine if there is a relationship between two variables that cannot be explained by chance (i.e., the data are statistically significant). Later in this book, guidance is provided on how to find the appropriate statistic and how to employ it in the examination of your data.

One of the issues you will encounter in both explanatory and evaluative research is whether your data can be explained by chance. Statistical significance refers to the likelihood that a given set of data would occur by chance. You will see much on this theme later. If your data can be explained by chance, you cannot logically conclude that the data are meaningful. For example, if you find that your clients’ posttest scores for anxiety are better than their pretest scores but you find that the data failed to be statistically significant, you cannot logically declare that you found that the clients gained with regard to reduced anxiety. You cannot do this because you found data that can be explained by chance, so the next time you do the same study, you are just as likely to find that they did not gain as to find that they did. Your data must do better than being the result of chance to be taken seriously.

Social Work Research Methods

Подняться наверх