BulletPointsummary per chapter with the 15th edition of The Practice of Social Research by Babbie - Chapter

How are human inquiry and science conducted? - BulletPoints 1

• Our attempts to learn about the world are only partly linked to direct observations, but in a larger part it comes from agreed-on knowledge that others give to us. Two important sources of agreed-on knowledge are tradition and authority.
• We distinguish four common errors that are made during casual day-to-day inquiries: inaccurate observations, overgeneralisation, selective observation and illogical reasoning.
• Social research has three main purposes: exploring, describing and explaining. Many social research projects reflect more than one of these purposes. 
• Inductive theories reason from specific observations to general patterns. Deductive theories start from general statements and predict specific observations.
• Tolerance for ambiguity refers to the ability to hold conflicting ideas in your mind simultaneously, without denying or dismissing any of them. Tolerance for ambiguity is an important skill in social research.
• Quantitative data are numerical, qualitative data are not numerical.
• Most organised research begins with a description of what is planned in the project, what questions it will raise and how it will answer them. Often, such research proposals are also created for the purpose of getting the resources needed to conduct the research envisioned. 

How are paradigms and theories used within the field of social research? - BulletPoints 2

• A paradigm is a fundamental model or scheme that organises our view of something. Social scientists use a variety of paradigms to organise how they understand and inquire into social life. Paradigms underlie social theories and inquiry. Whereas theories seek to explain, paradigms provide ways of looking.
• We can make a distinction between different types of theories: macro theories and micro theories.
• Social scientists have formulated several paradigms for understanding social behavior. We distinguish several of the central paradigms in our society: the positivistic paradigm, the conflict paradigm, the symbolic interactionistic paradigm, the structural functionalist paradigm, ethnomethodology, feminist paradigms, critical race theory, and rational objectivity. This rich variety of theoretical paradigms can be brought to bear on the study of social life. With each of these fundamental frames of reference, useful theories can be constructed.
• The elements of social theory include observations, facts, laws, theories, concepts, variables, axioms or postulates, propositions and hypotheses.
• Through operationalisation, researchers develop operational definitions or specify exact operations involved in measuring a variable. Hypothesis testing consists of making observations that are aimed at finding out whether the formulated hypothesis accurately describes reality. Aside of the hypothesis, researchers also tend to formulate a null hypothesis, which is the hypothesis that suggests that there is no relationship among the variables under study.
• Social science theory and research are linked through two logical methods: Deduction and induction.
• Science is a process involving an alternation of deduction and induction. We make a distinction between deductive theory construction and inductive theory construction.
• There is no simple recipe for conducting social science research. It is far more open-ended than the traditional view of science suggests.
• Ultimately, science depends on two categories of activity: logic and observation.

In what way do ethics and politics play a role within the field of social research? - BulletPoints 3

• Within the conduction of social research, ethical considerations such as voluntary participation and no harm to the participants are extremely important core values.
• The informed consent ensures that subjects must base their voluntary participation in research projects on a full understanding of the possible risks involved.
• Social researchers have ethical obligations to the community of researchers as well as to subjects. These obligations include reporting results fully and accurately, as well as disclosing errors, limitations and other shortcomings of the research.
• Anonymity refers to the sit-in in which even the researcher cannot link specific information to the individuals it describes. On the other hand, confidentiality refers to the situation in which the researcher promises to keep information about subjects private.
• Institutional review boards (IRBs) review research proposals involving human subjects so that they can guarantee that the subjects' rights and interests will be protected. The chief responsibility of an IRB is to ensure that the risks faced by human participants in research are minimal.
• Professional associations in several disciplines publish codes of ethics to guide researchers. These codes of ethics describe what is considered acceptable and unacceptable professional behavior.
• Laud Humphrey's research provoked considerable controversy both inside and outside the social science community. Some critics charged Humphreys with a gross invasion of privacy in the name of science, whereas others were mostly concerned about the deceit involved.
• Milgram's experiments have been criticized both methodologically as well as ethically. On the ethical side, critics have particularly cited the detrimental effects of the experiment on the participants.
• Although most researchers agree that political orientation should not unduly influence research, in practice it can be very difficult to separate politics and ideology from the conduct of research. However, although researcher should not let their own values interfere with the quality and honesty of their research, this does not mean that researchers cannot or should not participate in public debates and express both their scientific expertise and personal values. 

How are research projects designed? - BulletPoints 4

• We distinguish the following three principal purposes of social research: exploration, description and explanation. Research studies often combine more than one purpose
• Idiographic explanations seek an exhaustive understanding of the causes producing events and situations in a single or limited number of cases. The nomothetic model aims at a general understanding, not necessarily a complete one, of a class of phenomena, by using a small number of relevant causal factors. 
• When social researchers say there is a causal relationship between two variables, they mean that there is a statistical correlation between the variables, the causal variable occurs earlier in time than the affected variable, and the observed effect cannot be explained the observed correlation as spurious.
• When social researchers say that one variable causes another, they do not necessarily mean to suggest complete causation, to account for exceptional cases, or to claim that the causation exists in a majority of cases.
• A necessary cause is one that must be present for the effect to occur. A sufficient cause is one that will always produce the effect in question.
• We distinguish individuals, groups, organisations, social interactions and social artifacts as the most important and most common units of analysis in social research.
• The ecological fallacy refers to the assumption that something that is learned about an ecological unit says something about the individuals making up that unit. With the ecological fallacy, we confuse units of analysis in such a way that we base conclusions about individuals solely on the observation of groups.
• Reductionism refers to the attempt to understand a complex phenomenon in terms of a narrow set of concepts.
• Longitudinal studies have an obvious advantage over cross-sectional studies in providing information describing processes over time. But this advantage often comes at a heavy cost in both time and money, especially in a large-scale survey.
• Most studies use a single method for collecting data, such as in the case of surveys, experiments and field researches. Using more than one method in a given study can however yield a more comprehensive understanding.
• Research design involves a set of decisions regarding what topic is to be studied, among what population, with what research methods and for what purpose. 
• We distinguish several key elements that should be included in a research proposal: the problem or objective, a literature review, the subjects for study, measurement, data-collection methods, analysis, schedule, budget and depending on the nature of the research, an ethical approval of the institutional review board.

How do conceptualisation, operationalisation and measurement take place? - BulletPoints 5

• The definition of terms and variables is critical in social research. We make a distinction between conceptions and concepts.
• Conceptualisation refers to the process of specifying observations and measurements that give concepts definite meaning for the purposes of a research study.
• We make a distinction between real, nominal and operational definitions.
• The impact of wording choices is a never-ending concern for social researchers, and the testing of different words will never end. 
• Operationalisation is an extension of conceptualisation that specifies the exact procedures that will be used to measure the attributes of variables. Operationalisation involves a series of interrelated choices: (1) specifying the range of variation that is appropriate for the purposes of a study, (2) determining how precisely to measure variables, (3) accounting for relevant dimensions of variables, (4) clearly defining the attributes of variables and their relationships, (5) and deciding on an appropriate level of measurement.
• Researchers must choose from four types of measures that capture increasing amount of information: nominal, ordinal, interval and ratio. The most appropriate level depends on the purpose of the measurement.
• We distinguish different criteria of the quality of measures: precision, accuracy, reliability and validity.
• Reliability refers to getting consistent results form the same measure. Researchers can test or improve the reliability of measures through the test-retest method, the split-half method, the use of established measures and the examination of work performed by research workers.
• Validity refers to getting results that accurately reflect the concept being measured. The yardsticks for assessing a measure's validity include face validity, criterion-related validity, construct validity and content validity. 
• Creating specific, reliable measures often seems to diminish the richness of meaning our general concepts have. This problem is inevitable. The best solution is to use several different measures, tapping the various aspects of a concept.

How do researchers apply indexes, scales and typologies within the field of social research? - BulletPoints 6

• Single indicators of variables seldom capture all the dimensions of a concept, have sufficient validity to warrant their use, or permit the desired range of variation to allow ordinal rankings. Composite measures, such as scales and indexes, solve these problems by including several indicators of a variable in one summary measure.
• An index is a type of composite measure that summarises and rank-orders several specific observations and represents some more general dimension. A scale is a type of composite measure composed of several items that have a logical or empirical structure among them.
• The first step in creating an index is selecting items for a composite index, which is created to measure some variable. Criteria of item selection include face validity, unidimensionality, the degree of specificity with which a dimension is to be measured, and the amount of variance provided by the items.
• The second step in index construction is to examine the empirical relationships among the items being considered for inclusion. If different items are indeed indicators of the same variable, then they should be related empirically to one another. In constructing an index, the researcher needs to examine bivariate and multivariate relationships among the items. 
• The third step in the process of index construction is index scoring. With index scoring, the researcher assigns scores for particular responses, thereby creating a single composite index out of the several items. Index scoring involves deciding the desirable range of scores and determining whether items will have equal or different weights.
• The fourth step in index construction is to handle missing data. Researchers can apply various techniques that allow items to be used in an index in spite of missing data.
• The last step in the process of index construction is index validation. We make a distinction between internal validation and external validation. 
• We distinguish four types of scaling techniques: the Bogardus social distance scale, Thurstone scaling, Likert scaling, semantic differntial and Guttman scaling.
• A typology is a nominal composite measure often used in social research. Typologies can be used effectively as independent variables, but interpretation is difficult when they are used as dependent variables. 

What is the logic behind sampling? - BulletPoints 7

• Social researchers must select observations that will allow them to generalise to people and events not observed. Often, this involves the process of sampling, which is the selection of people that will be observed. Understanding the logic of sampling is essential to doing social research. We make a distinction between probability sampling and non probability sampling.
• Non-probability sampling techniques are techniques in which samples are selected in some way not suggested by probability theories. Non-probability sampling methods include reliance on available subjects, purposive sampling, snowball sampling and quota sampling. In addition, researchers studying a social group may make use of informant. Each of these techniques has its uses, but none of them ensures that the resulting sample will be representative of the population being sampled.
• The fundamental idea behind probability sampling is that in order to provide useful descriptions of the total population, a sample of individuals from a population must contain essentially the same variations that exist in the population.
• Representativeness refers to the degree to which a sample has the same distribution of characteristics as the population from which it was selected. Random selection is often a key element in probability sampling. In random selection, each element has an equal chance of being selected independently of any other event in the selection process.
• A sample's representativeness depends directly on the extent to which a sampling frame contains all the members of the total population that the sample is intended to represent.
• We distinguish several sampling designs that are available to researchers: simple random sampling, systematic sampling and stratification.
• Multistage cluster sampling is a relatively complex sampling technique that is frequently used when a list of all the members of a population does not exist. The general guideline for cluster design is to maximise the number of clusters selected while decreasing the number of elements within each cluster. The efficiency of cluster sampling is based on the ability to minimise the listing of population elements. 
• Probability proportionate to size sampling (PPS) is a special, efficient method for multistage cluster sampling. It is a type of multistage cluster sampling in which clusters are selected, not with equal probabilities (like with EPSEM), but with probabilities proportionate to their sizes.
• Weighting refers to the process of assigning these weights to cases that were selected into a sample with different probabilities of selection. In the simplest scenario, each case is given a weight equal to the inverse of its probability of selection.

How do researchers conduct experiments? - BulletPoints 8

• Experiments provide an excellent vehicle for the controlled testing of causal processes.
• The classical experiment tests the effect of an experimental stimulus, which is the independent variable, on a dependent variable. This is done through the pretesting and post testing of experimental and control groups. A classical experiment involves these three major pairs of components: independent and dependent variables, pretesting and post testing, and experimental and control groups.
• A double-blind experiment guards against experimenter bias because neither the experimenter nor the subject knows which subjects are in the control and experimental groups.
• Ideally, the control group represents what the experimental group would have been like if it had not been exposed to the experimental stimulus. Probability sampling, randomisation and matching are all methods of achieving comparability in the experimental and control groups.
• Campbell and Stanley describe three forms of pre-experiments: the one-shot case study, the one-group pretest-posttest design, and the static-group comparison. Preexperimental designs are experimental designs that do not meet the scientific standards of experimental designs and are used because the conditions for full-fledged experiments are impossible to meet. 
• Campbell and Stanley list, among others, eight sources of internal invalidity in experimental designs: history, maturation, testing, instrumentation, statistical regression, selection biases, experimental mortality and demoralisation. Internal invalidity refers to the possibility that the conclusions drawn from experimental results may not accurately reflect what has gone on in the experiment itself. The threat of internal invalidity is present whenever anything other than the experimental stimulus can affect the dependent variable. The classical experiment with random assignment of subjects guards against each of these sources of internal invalidity.
• The Solomon four-group design not only rules out interactions between testing and the stimulus, it also provides data for comparisons that will reveal the amount of such interaction that occurs in the classical experimental design. This knowledge allows a researcher to review and evaluate the value of any prior research that used the simpler design.
• Field experiments are formal experiments conducted outside the laboratory, in a natural setting. In a controlled field experiment, researchers exposed the Pygmalion effect as one phenomenon that researchers must account for in their experimental design. The Pygmalion effect refers to the tendency to see in others what we have been led to expect.
• Natural experiments often occur in the course of social life in the real world. Because in natural experiments the researcher must take things pretty much as they occur, such experiments raise many of the validity problems discussed earlier.
• The primary weakness of experiments is artificiality: What happens in an experiment may not reflect what happens in the outside world. The strengths of experiments include the isolation of the independent variable, which permits causal inferences, the relative ease of replication, and scientific rigor. 
• It is important that researchers determine whether a particular deception is essential to their experiment and whether the value of what may be learned form the experiment justifies this ethical violation.

How do researchers conduct survey research? - BulletPoints 9

• Survey research refers to the administration of questionnaires to a sample of respondents that is selected form some population. Survey research is especially appropriate for making descriptive studies of large populations. Survey data may be used for explanatory purposes as well.
• Items in a questionnaire should observe several guidelines.
• Questionnaires can be administered in four basic ways: self-administered paper questionnaires, face-to-face interviews, telephone surveys, or online surveys. 
• Response rates are important, since the nonrespondents may differ significantly from respondents, resulting in the survey results giving an inaccurate picture of the population under study.
• The essential characteristic of interviewers is that they be neutral; their presence in the data-collection process must not have any effect on the responses given to the questionnaire items.
• Increasingly, researchers are combining different modes in a single survey, benefiting from the advantages of each. The best combination of modes depends on the respondents and the topic and format of the survey.
• Survey research in general offers advantages in terms of economy, the amount of data that can be collected, and the chance to sample a large population.  Survey research has the weaknesses of being somewhat artificial, potentially superficial, and relatively inflexible. It is difficult to use surveys to gain a full sense of social processes in their natural settings.
• Secondary analysis provides social researchers with an important option for collecting data cheaply and easily, but at a potential cost in validity.
• Surveys often ask for private information, which researchers must keep confidential. Aside of this, because asking questions can cause psychological discomfort or harm to respondents, the researcher should minimise this risk.

How do researchers conduct qualitative field research? - BulletPoints 10

• Field research involves the direct observation of social phenomena in their natural settings. We do field research whenever we observe or participate in social behavior and try to understand it.
• John Lofland and his colleagues have discussed several elements of social life that are appropriate for field research. Among these topics are practices, episodes, encounters, roles, relationships, groups, organisations, settlements and lifestyles or subcultures.
• When you use field research methods, you are confronted with decisions about the role you will play as an observer and your relations with the people you are observing. In field research, observers can play any of several roles, including participating in what they want to observe. 
• Anthropologists sometimes use the term emic perspective in reference to taking on the point of view of those being studied. In contrast, the etic perspective maintains a distance from the native point of view in the interest of achieving more objectivity. Social researchers often refer to the concerns of balancing between these two perspectives as a matter of reflexivity. 
• Field research can be guided by any one of several paradigms, such as naturalism, ethnomethodology, grounded theory, case studies and the extended case method, institutional ethnography, and participatory action research (PAR).
• The process of conducting field research is characterised by preparing for the field, qualitative interviewing, conducting focus groups and recording observations.
• Steiner Kvale has identified seven stages in the complete interviewing process: thematising, designing, interviewing, transcribing, analysing, verifying, and reporting.
• Compared with surveys and experiments, field research measurements generally have more validity but less reliability. 
• Conducting field research responsibly involves confronting several ethical issues that arise form the research's direct contact with subjects.

What does unobtrusive research entail and how is it conducted? - BulletPoints 11

• Content analysis is a social research method appropriate for studying human communications. Researchers can use it to study not only communication processes but other aspects of social behavior as well. It is the study of recorded human communications, such as books, websites, paintings and laws.
• Coding is the process of transforming raw data into categories based on some conceptual scheme. Coding may attent to both manifest and latent content.
• The advantages of content analysis include economy, safety and the ability to study processes occurring over a long time. Also, it is unobtrusive. Content analysis is however also characterised by some disadvantages.
• A variety of government and nongovernment agencies provide aggregate statistical data for studying aspects of social life. This method is particularly significant because existing statistics should always be considered as at least a supplemental source of data. Existing statistics can often provide a historical or conceptual context within which researchers locate their original research. Existing statistics can also provide the main data for a social science inquiry.
• Social scientists can also use comparative and historical methods. These are aimed to discover patterns in the histories of different cultures. It involves the use of historical methods by sociologists, political scientists, and other social scientists to examine societies over time and in comparison with one another
• Big data refers to the gigantic data sets being automatically compiled from online activity. This is very much a method under development at present.
• Sometimes even unobtrusive measures can raise the possibility of violating subjects' privacy. The general principle of honest observation, analysis and reporting apply to all research techniques.

How do researchers conduct evaluation research? - BulletPoints 12

• Evaluation research is a form of applied research that studies the effects of social interventions. Evaluation research refers to a research purpose rather than a specific research method. This purpose is to evaluate the impact of social interventions.
• A social intervention is an action taken within a social context for the purpose of producing some intended results.
• The topics as well as the intent of evaluation research takes many forms. We distinguish needs assessment studies, cost-benefits studies and monitoring studies.
• To conduct evaluation research, we must be able to operationalise, observe, and recognise the presence or absence of what is under study.
• We distinguish three main types of research designs that are appropriate for evaluation research: experimental designs, quasi-experimental designs and qualitative evaluations. Evaluation researchers typically use experimental or quasi-experimental designs.
• Social indicators can provide an understanding of broad social processes. Social indicators refer to measurements that reflect the quality or nature of social life. Computer-simulation models hold the promise of allowing researchers to study the possible results of social interventions without having to incur those results in real life.
• Evaluation research entails special ethical problems as it is embedded in the day-to-day events of real life. Evaluation research may bring added pressure to produce specific results, as desired by interested parties. Unethical actions in an evaluation study can have more-severe consequences than such actions in other types of research. 

How is qualitative data analysis conducted? - BulletPoints 13

• Qualitative analysis is the non-numerical examination and interpretation of observations. The purpose of qualitative analysis is to discover the underlying meanings and patterns of relationships. Qualitative analysis involves a continual interplay between theory and analysis.
• For the most part, in examining your data you will look for patterns appearing across several observations that typically represent different cases under study. This approach is called cross-case analysis. Huberman and Miles offered two strategies for cross-case analysis: variable-oriented and case-oriented analysis.
• Examples of approaches to the discovery and explanation of patterns are Grounded Theory Method (GTM), semiotics, and conversation analysis (CA).
• The processing of qualitative data is as much art as science. We distinguish three key tools for preparing data for analysis: coding, demoing and concept mapping.
• Many computer programs, such as Qualrus and NVivo, are specifically designed to assits researchers in the analysis of qualitative data. Nowadays, qualitative data analysis (QDA) are great in number.
• Researchers need both qualitative and quantitative analysis for the fullest understanding of social science data. 
• Validity and reliability are ways to asses the quality of qualitative research.
• At least two ethical issues raise special concern in the analysis and reporting of qualitative research. The subjective element in qualitative data analysis provides an added challenge to avoiding bias in the interpretation of data. Because the qualitative data analyst knows the identity of subjects, taking special steps to protect their privacy is crucial.

How is quantitative data analysis conducted? - BulletPoints 14

• Quantitative analysis involves the techniques by which researchers convert data to a numerical form and subject it to statistical analyses.
• A codebook is the document that describes the identifiers assigned to different variables and the codes assigned to represent the attributes of those variables. A codebook serves two essential functions.
• Univariate analysis is the analysis of a single variable. Involves describing a case in terms of a single variable. Because univariate analysis does not involve the relationships between two or more variables, its purpose is descriptive rather than explanatory. Several techniques allow researchers to summarise their original data to make them more manageable while maintaining as much of the original detail as possible. Frequency distributions, averages, grouped data, and measures of dispersion are all ways of summarising data concerning a single variable. 
• Subgroup comparisons can be used to describe similarities and differences among subgroups with respect to some variable. Collapsing response categories and handling "don't knows" are two techniques for presenting and interpreting data. 
• Bivariate analysis focuses on relationships between variables rather than comparisons of groups. Bivariate analysis explores the statistical association between the independent variable and the dependent variable. Its purpose is usually explanatory rather than merely descriptive. 
• Multivariate analysis is a method for analysing the simultaneous relationships among several variables. It may also be used to understand the relationship between two variables more fully.
• Sociological diagnostics is a quantitative analysis technique for determining the nature of social problems such as ethnic or gender discrimination.
• Unbiased analysis and reporting is as much an ethical concern in quantitative analysis as it is in qualitative analysis. Subjects' privacy must be protected in qualitative data analysis and reporting.

What is the logic behind multivariate data analysis? - BulletPoints 15

• The elaboration model is a method for multivariate analysis appropriate for social research. It is primarily a logical model that can illustrate the basic logic of other multivariate methods. The elaboration model is a logical model for understanding the relationship between two variables by controlling for the effects of a third. 
• The logical relationships of the variables differ depending on whether the test variable is antecedent to the other two variables or intervening between them. The outcome of an elaboration analysis may be replication, explanation, interpretation or specification. 
• Ex post facto hypothesising, or the development of hypothesis "predicting" relationships that have already been observed, is invalid in science, because disconfirming such hypotheses is impossible.

What does the practice of social statistics look like? - BulletPoints 16

• Descriptive statistics are used to summarise data under study. The PRE model is a logical model for assessing the strength of a relationship by asking how much knowing values of one variable would reduce our errors in guessing values of the other. The measures of association that are based on the PRE model include lambda, gamma and Pearson's product-moment correlation.
• Regression analysis represents the relationships between variables in the form of equations, which can be used to predict the values of a dependent variable on the basis of values of one or more independent variables.
• We distinguish several different types of regression analyses: linear regression analysis, multiple regression analysis, partial regression analysis, and curvilinear regression analysis.
• Inferential statistics are used to estimate the generalisability of findings arrived at through the analysis of a sample to the larger population from which the sample was selected. Some inferential statistics estimate the single-variable characteristics of the population. Others, which are called tests of statistical significance, estimate the relationships between variables in the population.
• The level of significance of an observed association is reported in the form of the probability that the association could have been produced merely by sampling error. To say that an association is signifiant at the 0.05 level is to say that an association as large as the observed one could not be expected to result from sampling error more than 5 times out of a 100.
• Chi square is appropriate for testing the statistical association of relations found in typically nominal or ordinal tabular data. The t-test is a commonly used measure for judging the statistical significance of difference sin group means, and it is used for interval or ratio data.
• Statisticians speak of two different kinds of error: Type I error and Type II error.
• We distinguish a few principal reservations that we should have when using tests of statistical significance. 
• We distinguish several other multivariate techniques: path analysis, time-series analysis, factor analysis, analysis of variance (ANOVA), log-linear models, odds-ratio analysis, Geographic Information Systems (GIS), demographic analyses, et cetera.

What do the processes of reading and writing look like within the field of social research? - BulletPoints 17

• Meaningful scientific research is inextricably wed to communication: knowing how to read and write it requires practice. Social researchers can access many resources, including the library and the Internet, for organising a review of the literature that exists on a particular topic. Reading scholarly literature is different from reading other works, such as novels.
• Unless research is properly communicated, all the efforts devoted to the various procedures discussed throughout the summary of this book will go for naught.
• Plagiarism refers to the situation in which a researcher presents someone else's word or thoughts as though they were their own. It must be avoided at all times.
• A review of literature should not be biased toward a particular point of view. Research ethics are a fundamental component of social science, not just a nice afterthought.