social science statistics

social science statistics

The Importance and Application of Statistics in Social Science Research

1. Introduction to Statistics in Social Science

Reliance exclusively on case studies can be just as distortionary. The law of large numbers, of which statistical inference makes such frequent use, asserts that the average of a large number of independently selected items, all of which are themselves averages of a large number of observations, will be more representative of the population from which these items were selected or the longer list of elements with which the average was constructed. It is precisely this property of averages that provides a strong base for making generalizations to the entire population from which the given sample was drawn. The task of the researcher is to select a sample in such way the results obtained by statistical inferences are not vitiated by the probability of random fluctuations.

Statistics must play an increasingly important role in all phases of social science research. This does not mean, however, that statistics should become a substitute for the social scientists or that the latter should close his eyes to the possible shortcomings of the statistical approach. The ultimate goal of the social sciences is to comprehend the behavior of men within each of the specific areas of social organization. This comprehensive understanding comes only when the social scientist has been able to pool together the various known relationships about the subject and is able to specify the conditions that control the integrated behavior of the social system. Often the discovery of new relationships is possible. It is only as a result of a process of scientific inquiry that the discoveries are made and new understanding appears.

2. Key Statistical Concepts and Methods for Social Science Research

Population is the entire world or universe, such as living things, chemicals, etc. A sample is a portion of the population. We usually want to generalize our inferences, so we draw a sample from the population. The data actually collected is called a sample.

2.1 Importance of Understanding Sample Distributions Understanding probability distribution is important for social science research in order to infer the characteristics of the population. No research can study the entire population because of the following problems: time, human resources, financial resources, communication, and transportation. To overcome these problems, most social science research tries to understand the sample distribution and infer the population distribution from the sample distribution. Thus, we can think of statistical analysis as a process of estimating the sample distribution using the sample data, or as a process of drawing conclusions from the sample distribution created using the sample data. The purpose of statistical analysis is to estimate the behavior of a sample from which the sample is drawn.

This section is designed to give readers an understanding of the key statistical concepts and the methodology employed in social science research. It is expected that readers will have a basic knowledge of research approach and research methods, including how to design questionnaires or interview guides, and how to analyze qualitative data. If that is not the case, we strongly recommend that readers first take the previous E-module on Introduction to Research, which is available from Okinawa. The key statistical concepts and methods are presented with the purpose of widening the scope of understanding the strengths and limitations of the method.

3. Data Collection and Analysis Techniques in Social Science

Infuriating as it may be, it seems that we use the developed highly mathematical machinery to facilitate obtaining a wish fulfillment that we would have even without it. Is the way we meet many of these ideas in our study of social behavior nowadays, which is simply to obtain statements of the obvious. The aim is to bring to the fore the ideas that we consider fundamental, to present them clearly, and develop the implications that logically follow from them. Theories are creative, and practicing statistics means translating these into soundly designed and meaningful experiments or observational studies. It is the view that everyone who wants to question what is done, and to play a responsibility for decisions that affect others must be to understand what is happening in the area in which he or she has an input. This applies to potential users as well as to practitioners.

My philosophy is that everyone will use statistics in his or her own field and it is obvious that a scholar should know the essentials of the subjects. It is therefore paradoxical not to assume that we have different needs and different backgrounds. Moreover, in our increasingly technological world, decision-makers in any field need to measure uncertainties in the data available and to project them into the future. Indeed, statistical thinking is a very special case of human thought. It seems to be embedded in the evolutionary message that we must react to the message of our senses. Once this is accepted, we can legitimize the search for a few general principles transcending specific data.

2. The Role of Statistics in Social Sciences

This chapter introduces the importance of statistics in the field of social sciences. It examines the body of knowledge informing social science research and identifies how this can be used effectively in the design and analyses of research studies reported in the literature, both descriptive and inferential. The chapter then explores mechanisms for assembling information from social science research and presents the design and manipulation of research data collected when answering specific research questions in support of the data creation, exploration, presentation, and examination process that appears to be central to the use of package statistical programs. Finally, data transformations are discussed that are key issues in social science research. All these are important issues in that they assist in the organization of ideas and in fostering a culture of scholarly research.

1. Introduction

4. Ethical Considerations in Using Statistics in Social Science Research

A third ethical consideration is the necessity for clear and transparent communication. It has been estimated that less than 1% of the technical findings in applied social science research are actually read by people outside the scientific community. However, methodologists often hold a powerful role in terms of determining what is published and what is not published because of both the journal reviewing process and the process of peer evaluation. The freedom that researchers have is not the freedom to express themselves in technical language but to make themselves understandable. A final ethical consideration is the necessity for accountability to consumers. Seekers of patent medicines were not the most literate members of the populace, yet it became evident to the individuals who studied their problems that those researchers could not evade receiving the benefit or the odium of their labor. Any piece of research that is intended to advance understanding usually does have consumers who are affected by the production of this research. The effects may be beneficial or they may be harmful, but that cannot be the point. If the researcher wishes to defend his/her work by saying, “I couldn’t have anticipated how it would have been received or used,” he/she can be criticized for irresponsibility.

A second important ethical consideration in using statistics in social science is the relevance of findings. In the search for prestige, status, and power, researchers sometimes lose sight of the larger purpose of science. In the field of social science, that purpose is to contribute to the understanding of the behavior of human beings and companies. However, the fruits of many researchers’ labor are complex data exercises and sophisticated statistical tests that add only slightly to such understanding. In such situations, the focus is misplaced. These findings could have a special resonance in social science research because of their potential to influence policies, procedures, and practices that can affect the well-being of individuals and companies. For this reason, one must question the value of “using statistical analyses to answer questions that nobody cares about.” What one needs to do is to develop measures and methods that expand our understanding of the events and processes that one hopes to capture.

The first ethical issue surrounding the use of statistics is the integrity of the information used in statistical analyses. The information used in statistical analyses is usually generated through sampling procedures. The integrity of a statistical analysis depends on the honesty with which data are acquired. Given that most empirical research is based on the use of samples, the use of statistical procedures would be non-informative if the samples used in analyses are not reflective of the population from which they were selected. Since most research in social science relies on this procedure, the findings from statistical analyses should only be expected to apply to the populations sampled if the sample is, in fact, an unbiased sample. One manner in which one could ensure that the sample is an unbiased sample and is therefore truly reflective of the population is to clearly stipulate the sampling procedure used and the properties of the population that the sample is likely to represent. Another manner to assess the integrity of the sample is to engage in a thorough examination of evidence that would lead to a plausible validation of the samples that were selected.

There are a number of ethical considerations that should be taken into account when using statistics in social science research. Although much attention is devoted to the technical aspects of statistics, the use of statistics in social science raises a number of important ethical questions.