What research method is most commonly used in social science?
#1 – Introduction
- Why are you taking this research methods course here & now?
- Required by my school/department? – “I’m forced!!”
- My own interest?
- Need scientific ways to gather, analyze, & interpret data? ….
- Other reasons –
- Conducting my own research – asking accurate questions, matching research questions to sound methods, following correct procedures, analyzing data scientifically, reaching logical conclusions …;
- Reading, understanding, & evaluating other people’s studies;
- Gathering & evaluating info in daily life – be an educated consumer of information
- Be a more knowledgeable ‘Reflective Practitioner’?
- Nature of Social Research
- Process of Social Research
- Theory and Research
- Influences on Social Research
- Research Strategies
- Criteria in Social Research
- Research Designs
Nature of Social Research
- What is social research? – academic research on topics relating to questions relevant to the social scientific fields.
- Why do social research? – “curious George”, there is an aspect of our understanding of what goes on in life, or
in society, that is to some extent unresolved.
- a gap, or an unresolved issue, in the literature;
- an inconsistency between several studies;
- development and/or new phenomena in society;
- To acquire knowledge for the sake of practice
- Why is it important to study methods?
- sensitizes you to the choices available to social researchers;
- provides you with an awareness of the “dos” & “don’ts” when employing a particular approach to collecting or analyzing data;
- provides: insights into overall research process, awareness of what constitutes good & poor research;
- transfers your learned skills to other areas and to explore new skills;
- exposes you to various real-life research examples.
Process of Social Research
- The Scientific Method
- Literature review
- Concepts & theories
- Research questions – hypothesis
- Sampling cases
- Data collection
- Data analysis
- Writing up
Theory and Research
- Theory à Observations/Findings: based on what is known about & theoretical considerations of a particular domain, hypothesis/hypotheses is/are proposed & translated into operational terms, then, empirical analyses follows …
- Observations/Findings à Theory: theory is the outcome of research.
- Iterative ( ): a weaving back and forth between data and theory.
Influences on Social Research
|Epistemological considerations-theory of knowledge, esp. with regard to its methods, validity, and scope
|Positivism||Research on social sciences should imitate/apply the same research methods of natural sciences..|
|Principles: phenomenalism deductivism, inductivism objective, testable and refutable hypotheses.|
|Interpretivism||The subject matter of the social sciences-people and their institution- is fundamentally different from that of the natural sciences: social reality has a meaning for human beings, therefore human action is meaningful.|
|Study of the social world, requires a different logic of research procedure, reflecting the distinctiveness of humans and interpreting their actions and social world.|
|Ontological considerations-nature of being/social entities|
|Objectivism||social phenomena and their meanings exist as external facts independent of social actors/beyond our reach or influence|
|for example, organization, culture:
something external to the employees with almost tangible/objective reality of its own
|Constructionism/Constructivism||Social phenomena and their meanings are continually being accomplished by social actors: produced through social interaction and in a constant state of revision.
|Value: Personal Beliefs or Feelings of a Researcher|
|Different approaches||1. Suppressing values when conducting research- however, it is not feasible to keep a researcher\’s values totally in check, values can materialize at any point during the course of research.|
|2. Recognizing and acknowledging- research cannot be value free, but ensuring- there is no uncontrolled incursion of values in the research process and to be self-reflective.|
|3. Doing consciously value-laden research- some feminists writers: only research on women that is intended for women will be consistent with the wide political needs of women.|
|Practical considerations||Choice of research strategy, design, or method have to be closely linked wit the specific research question being investigate.|
|No prior research- exploratory, qualitative approach.|
|Nature of the topic and of the people being investigated (eg illicit activities.
Social Research Strategy
Research Strategy: a general orientation to the conduct of research.
|Emphasizes quantification in the collection and analysis of data; statistical analysis for summary and interpretations|
|Entails a deductive approach to the relationship between theory and research, in which the focus is placed and the testing of theories;||Predominantly emphasizes an inductive approach to the relationship between theory and research, in which the emphasis is placed on the generation of theories;|
|Has incorporated the practices and norms of the natural scientific model and of positivism in particular;||has rejected the practices and norms of the natural scientific model and of positivism in particular in preference for an emphasis on the ways in which individuals interpret their social world;|
|Embodies a view of social reality as an external, objective reality.||embodies a view of social reality as a constantly shifting emergent property of individuals’ creation.|
Social Research Strategy
|Fundamental differences- Quantitative vs. Qualitative research strategies|
|Principal orientation to the relationship between theory and research;||Deductive; testing of theory||inductive; generation of theory|
|Epistemological orientation||natural science model in particular positivism||interpretivism|
Mixed methods research: research that combines methods associated with both quantitative and qualitative research.
Criteria in Social Research
Validity: the degree to which the measurement process measures the variable that it claims to measure. (integrity of the conclusions generated from a research)
|Measurement validity/construct validity||Does a measure that is devised of a concept really reflect the concept that it is supposed to be denoting? (e.g., does the IQ test really measure variations in intelligence?)|
|Internal validity||Does a conclusion that incorporates a causal relationship between two or more variable hold?|
|if we suggest that x causes y, can we be sure that it is x that is responsible for variation in y and not something else that is producing an apparent causal relationship?|
|independent variable: the factor that has a causal impact (x) dependent variable: the effect (y)|
|External validity||can the results of a study be generalized beyond the specific research context?|
|Crucial: to generate representative samples when selecting participants in research.|
|Ecological validity||Are social scientific findings applicable to people\’s everyday, natural social settings?|
|the more social scientist intervenes in natural settings or creates unnatural ones (e.g., a laboratory or a special room to carry out interviews), the more likely that findings will be ecologically invalid.|
|Qualitative research often involves a naturalistic stance ( ethnographic research, interview), making it stronger that quantitative investigations in terms of ecological validity.|
Reliability (信度): are the results of a study repeatable? Are the measures devised for concepts in social sciences consistent / stable?
- Particularly at issue with quantitative research: if the same individuals’ IQ scores were wildly different when administered on 2 or more occasions, then we would consider it an unreliable measure.
- To assess the reliability of a measure of a concept, the procedures that constitute that measure must be replicable by someone else.
- Replication/replicability: an investigation’s capacity to be replicated.
Types of reliability:
▶ Inter-rater reliability (評分者信度) : degree of agreement between 2 observers who simultaneously record measurements.
▶ Test-retest reliability (再測信度) : comparing the scores obtained from 2 successive measurements of the same individuals & calculating a correlation between the 2 sets of scores.
▶ Internal Consistency reliability (內部一致性信度) : it is important to assess whether the various items that make up the measure are internally consistent. It assumes that the instrument contains multiple items, each of which is scored and combined together to generate an overall scores. Basically it is checked by Cronbach’s alpha value.
The relationship between reliability & validity
▶ Reliability is a prerequisite for validity – a measurement procedure cannot be valid
unless it is reliable: no reliability à no validity.
▶ high reliability (consistency of measurement): no guarantee of validity.
Research design: a framework to guide the execution of a research method and the collection & analysis of data.
- A choice of research design reflects decisions about the priority given to a range of dimensions of the research process:
- Expressing causal connections between variables (characteristics or conditions that change or have different values for different cases);
- Generalizing to larger groups of individuals;
- Understanding behavior & its meaning in specific social context;
- Appreciating social phenomena & their interconnections over time.
Research method: a technique for collecting data – self-completion questionnaire, structured interview schedule, participant observation
- Goal: to establish the existence of a cause-and-effect relationship between two (2) variables
- 4 basic of an experiment:
- Independent variable: the variable manipulated by the researcher.
- Dependent variable: the variable that is observed & measured for changes to assess the effects of manipulating the independent variable.
- Robust causal findings à very strong internal validity.
- Manipulation: manipulating the independent variable in order to determine whether it does in fact have an influence on the dependent variable.
- Participants are assigned to one of two or more experimental groups, each of which represents different types/levels of the independent variable; à
- To establish how far differences between the groups are responsible for variations in the dependent variable.
- However, vast majority of independent variables with which social researchers are concerned cannot be manipulated.
【明報文章】進入數碼時代，全球新聞行業一直在尋找有效的經營模式。近年，歐美新聞業的趨勢是重新確認向讀者收費的必要性。在香港，也有不少網媒主要通過讀者捐款或訂閱來支持營運。民眾對網上新聞內容的付費意欲（willingness to pay），也成為不少學者關注的課題。
- 1. Experimental design
- 2. Cross Sectional design
- 3. Longitudinal design
- 4. Case Study
- 5. Comparative design
1. Experimental design
- Classical experiment design (randomized experiment, randomized controlled trial (RCT))
- 1). Experiment/treatment group vs. control group (receiving no treatment) &&
- 2). Random assignment to the groups à
- Comparing: difference in dependent variable between the groups was due to the experimental manipulation alone: X causes Y à strong internal validity.
- May be difficult to establish external validity, ecological validity.
- Laboratory experiment vs. field (real-life setting) experiment
- 1. Experimental design
- Quasi-experiments: studies that have certain characteristics of experimental designs but that do not fulfil all the internal validity requirements.
- Natural experiment: not possible to randomly assign participants.
- No artificial interventions à very strong ecological validity.
- Significance of experimental design:
- A true experiment: removes doubts about internal validity & reflects the considerable emphasis placed on the determination of causality in quantitative research à a benchmark to judge quantitative research.
|Collecting data on more than one case and at a single point in time to collect a body of quantitative/quantifiable data in connection with 2+ variables, which are then examined to detect patterns of association.|
|Survey research||a cross-sectional design-data are collected predominantly by questionnaire or structured interview.|
|Reliability and measurement validity||Primarily related to the quality of the measures used to tap the concepts the researcher is interested in.|
|Replicability||usually high, if the researcher spells out procedures for : selecting respondents; designing measures of concepts; administering research instruments(structured interview or self-completion questionnaire); analyzing data.|
|Internal validity;||weak-cannot determine causal relationships
|External validity||strong, when the sample from which data are collected has been randomly selected
|Ecological validity||May be jeopardized, the frequently used instruments (self-completion questionnaires, structured observations schedules) disrupt the natural setting.|
|Non-manipulable variables:||ethnicity, gender, smoker vs. non-smoker (profoundly unethical).
- Usually an extension of survey research within a cross-sectional design;
- A sample is surveyed and is surveyed again on at least one further occasion;
- Can allow some insight into the time order of variables may make causal inferences.
|Two types of longitudinal design||Panel study||a sample (often a randomly selected national one), is the focus of data collection on at least two (and often more) occasions.|
|data may be collected from different types of \”case\” individuals, households, organizations, schools.|
|Cohort study||focus of data collection: either an entire cohort of people or a random sample of them.|
|the cohort is made up of people who share a certain characteristic- all being born in the same week, all having a certain experience (being unemployed/getting married on a certain day or in the same week)|
Panel study vs. cohort study
|Similar features||Design structure: data are collected in at least two waves on the same variables on the same people.|
|Concerned with: illuminating social change and improving the understanding of causal influences over time.
|Important differences||Panel study-1). can distinguish between age effects and cohort effects (effects due to being born at a similar time), because participants are born at different times: 2). needs rules and how to handle new entrants and exists.
|Cohort study- can distinguish only ageing effects.|
|Similar problems:||Sample attrition through death, moving, participants\’ withdrawal at later stages of research;
|Few guidelines as to when is the best juncture to conduct further waves of date collection;|
|Poorly thought out and collecting large amounts of data with little apparent planning.|
|Panel conditioning effect can occur-continued participation in a longitudinal study effects how respondents behave.
4. Case study design:
- Detailed & intensive analysis of a single case;
- Concerned with the complexity & particular nature of the case in question.
- What is a case? – a community, a school, a family, an organization, an event, a person … the case of H.M.
- Restricted external validity: issue to generalize.
- How well can the finding generate a theory?
5. Comparative design:
- Studying two contrasting cases using more or less identical methods.
- e.g., cross-cultural or cross-national research: to seek explanation for similarities & differences, or to gain a greater awareness & a deeper understanding of social reality in different national contexts.
- Key: to allow the distinguishing characteristics of two or more cases to act as a springboard for theoretical reflections about contrasting findings