7400.685-080 - Research Methods in FCS
School of Family and Consumer Sciences
Instructor: David D. Witt, Ph.D.
Key Terms #1 for Research Methodology in General: Use each of the following terms in sentence that illustrates your understanding of the concept as characterized by the definition given.
1. Theoretical – a concept that is concerned with developing, exploring, or testing theories or ideas that researchers have about how social phenomena operate.
2. Empirical - descriptions of events based on observations and measurements of reality.
3. Probabilistic – theoretical inferences made in terms of the statistical likelihood of being true.
4. Causal relationship – an event condition or outcome that is dependent on the existence of a prior or condition which has to do with the idea of cause and effect
5. Descriptive studies are designed primarily to document what is going on or what exists.
6. Relational studies look at the relationships between two or more variables
7. Causal studies are designed to determine whether one or more variables cause or affect one or more outcome variables.
8. Cross-sectional studies - take measurements at a single point in time.
9. Longitudinal studies - take place over more than one point in time.
10. Repeated Measures – are used in studies in which the same measurement is taken at two or more points in time
11. Time Series Analyses – are used in studies in which the same measurement is taken over many points in time and will include additional data – such as the analyses that have established the trends in global warming over centuries.
12. A variable is any concept that will vary from one respondent to the next when measured. For example, the members of a sample of respondents will vary in terms of their height, weight, or gender.
13. An attribute is a specific value on a variable - the variable sex or gender, for example, has at least two attributes: male and female.
14. Attributes of variables must have two additional qualities to make them useful in research – they should be:
15. Mutually exclusive from each other – meaning that a single attribute covers only one characteristic of the variable – it almost goes without saying that if you if you are male, then you cannot also be female. However if asked about income and choices are a) $20,00 and $30,000 or $30,000 and $40,000 respondents could choose either if they make $30,000 since the income attributes aren’t mutually exclusive. Attributes also should be:
16. Exhaustive – meaning that ALL the possible attributes are present in the response set. For example, a) less than $20,000 b) $20,000 to $29,999 or C) over $30,000 are both mutually exclusive and exhaustive.
17. An Independent variable is a type of variable that “operates on its own” and is not influenced by other variables within the confines of the study. Typically these would be variables such as gender, age, or race.
18. A Dependent variable is one that is affected by independent variables, as when a certain attitude (dependent) is different for men versus women (gender as independent).
19. A Mediating variable is one that theoretically resided between independent and dependent variables. For example gender effects are mediated by age in predicting a specific attitude or behavior.
20. In constructing measureable variables for use in research, Finally, the attributes of a variable should be both exhaustive and mutually exclusive.
21. Relationship refers to the correspondence between two variables – how they are theoretically connected to each other. This is important in converting theories to testable hypotheses. There are a few types of relationships:
22. A Correlational relationship means that two variables operate in a predictable manner without having to note that one “causes” the other. Correlational relationships just mean when one variable exists with a specific attribute, it is likely that another variable will be present with a specific value.
23. A Causal relationship a correlational relationship WITH an expectation of causation by another variable.
24. Several terms describe the major different types of patterns of relationships:
a. No relationship at all
b.The positive relationship: high values on one variable are associated with high values on the other, and low values on one are associated with low values on the other
c. Negative relationship: high values on one variable are associated with low values on the other. This is also sometimes termed an inverse relationship
d.Curvilinear relationship: shows a relationship that changes over the range of both variables
25. Hypothesis - a specific statement of prediction that is derived from a theory. When “operationalized” in research, the hypothesis describes in concrete terms what the research expects to happen.
26. Alternative hypothesis - is the hypothesis that the researcher expects to find – the predictive hypothesis.
27. Null hypothesis – is the hypothesis that predicts the opposite of the alternative
28. One-tailed hypothesis -refers to predictions that assume directionality – as in the sentence, “Usually men will be taller than women.” Keep this in mind when reading the next definition.
29. Two-tailed hypothesis – refers to predictions that do not assume directionality, as in the sentence – “there will be a difference in height between men and women.”
30. Quantitative vs. qualitative Data: Quantitative data: is in numerical form .
Qualitative data: not in numerical form. Qualitative data could be much more than just words or text, such as photographs, videos, sound recordings, and so on
31. Unit of Analysis refers to the entity (individual, pair, or group) that is of interest to the researcher. It is important to keep the unit of analysis in mind when designing measures. For example, if interested in families at the unit of analysis, it makes little sense in focusing attention on individuals only.
32. Anonymity refers to the assurance that an individual’s responses can never be traced to them (Sometimes also referred to as Confidentiality).
33. Cross-sectional studies take place at a single point in time.
34. Deductive: Top-down reasoning that works from the more general to the more specific.
35. Empirical questions are those in which answers are provided by direct observations and measurements.
36. Hypothetico-deductive model is one in which the alternative and null hypotheses are tested – making the acceptance of one the automatic rejection of the other.
37. Informed consent is the accepted research policy that participants in a research project are completely informed of all procedures and risks he study might entail, and that participants must give their consent before participating.
38. Institutional review board (IRB) consists of a panel of colleagues who review the ethical implications of research proposals, with the institutional authority to approve or deny sponsorship of a project.
39. Validity is the best possible measurement of a given variable or inference. In other words, a numerical approximation of the likelihood the research is actually measuring the concept intended.
Some Key Terms #2 on Sampling: Same deal – use each term in a sentence that illustrates your understanding of the concept.
1. The Sample is the: actual units of analysis you select to participate in your study.
2. Sampling refers to the process of selecting a group of units from a population so that any findings from the sample will be generalizable to the population.
3. Sampling error is the percentage chance that results from a sample will deliver erroneous findings. For example, sampling error of 3% means that the research can be 97% certain that the findings from the sample will represent those of the entire population.
4. A Sampling frame is the list from which a researcher draws a sample. For example, the email address listing of everyone who works at the University could be such a list. The subscribers to a magazine, blog, or newsletter could also be used.
5. Simple random sampling is the straightforward method of drawing a sample from a population so that every member of the population has an equal chance (probability) of being selected.
6. External validity refers to the degree to which the conclusions in your study would hold for other persons in other places and at other times. Perhaps the only way to be confident in the external validity is to replicate the study in other places, with other samples, and different points in time.
7. Multistage sampling refers to a practice of combining several sampling techniques to take advantage of the superior qualities of each technique, or to bolster one technique’s weaknesses with the advantages of another. For example, faced with the challenge of sampling the entire population of a nation, a researcher could divide the nation into parts, or clusters, and then randomly select a sample of clusters – thus introducing random selection. From here, each cluster could be sampled with a less than random method such as setting up a survey station at county fairs, or sending poll-takers into a community in each cluster.
8. Nonprobability sampling is a method that forgoes random selection in favor of convenience. It is often used in the early stages of descriptive theory testing.
9. A Population is that entire group to which the researcher wishes to generalize. From the population, the researcher will select a sample.
10. A Population parameter is the mean or average on measures obtained if every member of the population were surveyed or observed. In social science, this very labor-intensive method is never used. Instead, researchers use statistics instead, and generalize means to the population.
11. Probability sampling is a method that uses some form of random selection and is central to the researcher’s ability to generalize findings back to the population from which the sample is drawn. Without random selection, statistics become less valid and less useful.
12. Proportional quota sampling: is a method in the research continues to add participants to a sample from subgroupings of the population until a specific sample size is achieved. Here the samplings from subgroups are monitored so that the number of participants in each is proportional to the subgroups relative size within the population.
13. Random selection is the process that assures that the different units in your population are selected by chance.
14. Snowball sampling allows for participants to identify and refer new participants to the researcher for inclusion in the study.
15. Mean and Standard Deviation are the statistics used to describe characteristics of the sample in terms of measurements made. Known also as measures of central tendency, these statistics are generalizable to the population as long as the sample was randomly drawn.