1) Theories are to concepts as hypotheses are to variables.
Hypotheses depend on tests that suggest multiple variables. On the other hand, theories are made up of concepts ("Theory and Practice of Social Research" 4).
2) Does a hypothesis propose a relationship between dimensions or variables?
Yes. A hypothesis depends on the link that exists between two or several variables. Whereby, a slight change in variable ‘A,' leads to a change in variable ‘B.' In this case, it can be almost predictable that if the first event occurs on the first variable, then we expect variable ‘B' to change automatically.
3) The categories of each variable should meet what two requirements?
The classes of every variable must be equally exclusive and exhaustive. Exclusive in the sense that any observation made must belong to a single category. They must also be exhaustive in the sense that all observations can be easily categorized.
4) What, if anything, is the problem with the following categories of the variable "Political Views": Liberal, Moderate? If there is a problem, how might you correct it?
In this case, there are only two political views, Moderate and Liberal. This is certainly not exhaustive as some individuals may be conservatives. To correct this, we add another category called Conservative. In addition to this, we can also add another called ‘None' for all those who are not into politics. This would be totally exhaustive.
5) What, if anything, is the problem with the following categories of the variable" Political Perspective": Liberal, Democrat, Republican, or Conservative? If there is a problem, how might you correct it?
In this case, there exist confusing terms. The Democrats may as well regard themselves as liberals, as they are known to be open-minded and welcome new ideas. The Republicans can also consider themselves as conservatives. Therefore this is not exclusive as there is more than one category for each observation. To solve this, two categories must be eliminated to avoid repetition. We either do away with both Liberal and Conservative and leave Democrat and Republican or vice versa.
6) Construct a hypothesis based on the deprivation theory of church involvement using level of education as your independent variable.
With respect to the deprivation theory, the participation of individuals would defer depending on the level of education. Highly educated/College individuals are less likely to be involved in church matters as they are usually much involved in academics. Whereas those with little or basic/primary education would be much involved. The sphere of life has more to offer to the former group as they have to cope with a lot in their level. The dependent variable (church involvement), is either positively or negatively affected by the independent variable.
7) List the categories of the variables you used to construct your hypothesis in response to the previous question.
Primary Education.
8) Construct potential hypotheses to relate the following concepts and identify the independent and dependent variable in each hypothesis. In addition, list the categories of each variable.
Age and health
Health and age are related variables in the sense that health depends on age. Younger people have a good health as compared to most of the elderly individuals. With an increase in age the health condition of an individual changes as the body weakens. Therefore health is dependent on age. Age is the independent variable while health is a dependent variable.
Race and attitude toward affirmative action
The attitude towards affirmative action is dependent on the race. Whites as a result of cultural heritage are more inclined to take on white collar jobs as opposed to other races in the past. In this case, Race is the independent variable.
Gender and income
Both variables are closely linked. In most cases, gender affects income. Males are more muscular than females. Therefore they are more inclined to do most of the manual tasks available which mean a high income. On the other hand, females are not accustomed to doing heavy or tiresome tasks. This translates to less revenue activity. In most countries due to cultural reasons, ladies stay at home to take care of children. Thus, income is the dependent variable in this case.
9) Which of the following is not a dependent variable: grade point average, church attendance, age, the number of children? Age is not a dependent variable; it is independent as it is the factor that may lead to either positive or negative result to any dependent variable.
10) Which of the following is not a variable: occupation, the amount of television viewing, female, education level? Occupation is not a variable as it does not have the capability to take two or more forms. All the other listed factors apart from occupation are variables since they can be categorized.
Identify the level of measurement of each of the following variables. (Questions 1-3).
1) A researcher measures respondents' attitudes towards premarital sex by asking the following question:" If a man and woman have sexual relations before marriage, do you think it is always wrong, almost always wrong, wrong only sometimes, or not wrong at all?"
This is under nominal measurement as the categories cannot be ranked; neither can they fall under ratios and intervals. They are simply name labels expressing the views of various people regarding the subject matter. If for example, one person says it is wrong, you cannot quantify by numerals how wrong it is. There is no absolute scale of magnitude that tries to categorize according to the degree or amount ("Data Levels and Measurement").
2) A researcher measures the amount of television viewing by asking the following question: "On an average day, how many hours do you watch television?" Respondents are then asked to fill in the actual number of hours in the space provided. The researcher's measurement will be ordinal. This is simply because the response has to be in terms of the actual number of hours. For instance, one may say six hours a day. In this case, he has quantified the outcome using numerals (Types of data and measurement). The researcher can easily rank the result based on the answers of the respondents.
3) A researcher measures marital status by asking respondents whether they are currently married, widowed, divorced, separated, or never married. This is under nominal level. It simply categorizes according to names. There are no quantities, intervals or ratios. The status must be one of the listed categories with no regard to magnitude or rank.
4) Classify the variables in questions 1 to 3 as either discrete or continuous. Premarital sex is not a discrete variable neither is it continuous. The amount of television is discrete because the respondents will have to give the exact number of hours spent in watching television. Marital status is neither continuous nor discrete. It is a categorical nominal variable.
Indicate whether the following statements are True or False.
1) Certain variables can be measured at both the nominal and ordinal levels. False. Measurement done on a variable cannot be both nominal and ordinal, as this is antagonistic.
2) You are invited to screen a new movie and then asked to rate it as either excellent, good, fair, or poor. The level of measurement is nominal because the ratings differ in name. False. The standard of measurement is not nominal because the score is based on rankings.
3) A researcher asks respondents to indicate the last four digits of their social security number. The level of measurement for this variable is interval/ratio because the distance between categories is meaningful. True. For respondents to get it right they must use the last four digits as a benchmark. One can also subtract four out of the total number of digits available to arrive at the set point.
Works Cited
"Data Levels and Measurement." Statistics Solutions, www.statisticssolutions.com/data-levels-and-measurement/.
"The Theory and Practice of Social Research." Preparing Data for Analysis, U.S. Dept. of Education, Office of Vocational and Adult Education, 1981, pp. 3-9.
www.mymarketresearchmethods.com/types-of-data-nominal-ordinal-interval-ratio/.
