210 likes | 343 Views
RESEARCH DESIGN. Science involves making observations and interpreting the observations. In research design the following questions have to be answered: Be clear about what you want to find out Find out the best way to do it. Types/Purposes of research. Exploration
E N D
RESEARCH DESIGN • Science involves making observations and interpreting the observations. • In research design the following questions have to be answered: • Be clear about what you want to find out • Find out the best way to do it
Types/Purposes of research Exploration • Familiarizes researcher with the issue he/she is supposed to investigate • Useful when study is new and does not have much literature written on it. • Often conducted via focus group discussions.
Reasons why exploratory research is pursued • Curiosity • Feasibility • Future study
Description • Describes characteristics of a population Explanation • Involves explaining why things are the way they are. Answers the WHY questions. Most studies contain the three types of research
Units of Analysis • Is the measure to which your study refers to– that is, the level of social life on which the research question is focused. Individual • Unit of analysis is the individual
Groups • Information collected about groups – families, schools, work organizations, cities, states, countries. • Sometimes you might collect information from an individual but which refers to a group. • Unit of analysis vs. Unit of observation
Social Artifacts • Includes any product of social being –books, paintings, newspapers etc.
Dangers of Unit of Analysis • Ecological Fallacy – drawing conclusions about individual-level processes from group-level data • Reductionism fallacy – Reducing complex reality to simple explanations.
Research Designs Experimental Research • Seeks causal explanations based on analysis of how a group of subjects react to some treatment. • Best for determining causality Non Experimental Research • Involves studying naturally occurring variation in the dependent and independent variable without any intervention by the researcher. An example is survey research.
Time Dimension in Non Experimental Research • Time dimension is important in determining causation. Cross-Sectional Research Design • Data are collected at one point in time • Not easy to deal with the problem of spuriousness. Statistical Control is used to reduce this problem.
Statistical control – holding a variable(s) constant so that the relationship between two or more other variables can be assessed without the influence of the control variable. • Confidence in causality is boosted by identifying the causal mechanism, e.g. identifying Intervening variables.
Other means to determine causality in cross-sectional research • The independent variable is fixed at some point prior to the variation in the dependent. • We believe the respondents will give us reliable reports of what happened to them at an earlier time. • Can use historical data to support findings • We know that cases were equivalent on the dependent variable prior to the treatment.
Longitudinal Designs • Data are collected at two or more points in time. • Study changes over time. Types of longitudinal of designs • Trend studies – examines changes within a population over time • Cohort studies – examines specific sub-populations as they change over time. • Panel – examines the same set of people each time.
Research Proposal • Layout of details of what you are planning on researching. Components of a research proposal • Problem statement • Literature review • Study subjects • Measurement
Data collection • Analysis • Timetable • Budget
Practice questions • True/False • Indicate whether the sentence or statement is true or false. ____ 1. We study people for the purpose of seeing how variables are related to one another. ____ 2. Social scientific theory aims at the determination of logical and persistent patterns of regularities in social life. ____ 3. Idiographic explanations provide partial explanations of a particular instance.
Sally begins to think about the classes she's enrolled in at her university. She notices that she speaks a lot in three of her classes and very little in two of her classes. She also notices that the classes in which she speaks have female instructors and that she rarely speaks in the two classes with male instructors. She begins to list all her college courses, the gender of the professor and whether or not she spoke in the class. She notices that she tends to speak in classes where the instructor is female and rarely speaks when the instructor is male. Sally is using a. . inductive reasoning b. . qualitative analysis c. . deductive reasoning d. . ordinary human inquiry e. . nomothetic reasoning
Which of the following is a list of variables? a. female, Jewish, educational level b. plumber, professor, dentist c. occupation, political party preference, birthrate d. 21, violent, social class e. dishonest, conservative, farmer
Which of the following outlines the steps in the traditional deductive research model? a theory construction, operationalization of concepts, development of a hypothesis, and collection of data b operationalization of concepts, theory construction, development of a hypothesis, and collection of data c development of a theoretical hypothesis, theory construction, operationalization of concepts, and collection of data d collection of empirical data, theory construction, operationalization of concepts, and development of a hypothesis e theory construction, development of a hypothesis, operationalization of concepts, and collection of data
Short Answer questions • Distinguish between experimental and non-experimental research • Briefly state the importance of conducting social research • “Explanations in social science rest on deterministic model” What does this statement mean?
Essay question 1. Contrast the inductive model of theory construction with deductive model. Give examples of research that use each of these models.