Föreläsning 2 - Forskningsdesign, endast engelska anteckningar. PDF

Preview

Summary

Forskningsdesign, endast engelska anteckningar....

Description

FÖRELÄSNING2, TISDAG 20191008 

What is research design? Broad - the whole package of research, everything related to research Narrow - primarily the setup of the data collection process 

A research design is… What to i want do understand? Causal connections between variables Generalization from sample to population Understanding phenomena in their specific context Determining how and why phenomena change Determines the method to collect the data  

Research Typologies -

-



The Big Three - Experimental: Measuring the effect of change in one or more variable. Small group of people of the population of interest. One control group (constant variables) and one with manipulated variables. (Low lighting example) hypothesis testing. - Survey: S  tandardized collection of information. Survey is a standardized form of questiones. NOT the same thing as a questionnaires. Can be interviews too. - Case Study: Detailed study/knowledge about a single case or a few. Using a lot of different methods, ex observations, interviews, documents. Triangulate the case. Other types - Cross-section - at one point in time - critic: what variable is actually affecting the result? - Longitudinal - over a period of time - advantage: we know how things change, cause of effects. - panel s tudy: often a random group of population, and data collected across a number of timepoints. - cohort study a group with a shared characteristic (date of birth, marriage etc.) across a number of timepoints - trend study: different individuals sampled from the same population (not the same people) and data collected across a number of timepoints  not looking for change in individuals  - Prospective - follow up (how change affects) perspective - critic: - Retrospective - look back (cancer example) what is the influencing factor - critic: no control - Comparative - two similar phenomena in different contexts - comparing two situations/cases contrasting - understanding the phenomena

Research strategies and methods Two major research strategies for how one should collect and analyse data - Quantitative - numerical information - theory testing: deductive - natural scientific model - social reality objective - large study sample: reliable representation - there is a common truth - Qualitative - text and language - phenomenological model: we construct our reality - small study sample: understand the world people see - a research project will usually follow one or another - alt. a mixed methods, more used now. use one to understand the subject even deeper - antagonistic

 Research methods Techniques for collecting and analysing data - Sampling: T  hree main aspects - Who: define a study population and a target population - How: random, purposive and convenience - How many: n  umber of participants - Data collection - Interviews - focus groups - Questionnaires - open or closed questions; tests and scales - Observations - content/frequency of behavior; tests, scales - Documents - death certificates; diaries - Secondary Data - studies already made 

Data Analysis Form of analysis tied to form of data - Statistical analysis (numerical) - description of variation - correlational analysis - do things vary together? - comparing groups - do they differ? - statistical significance - is this a chance finding? - Content analysis ( text or non-numerical) - description of context - what is said and done? surface process - what is meant? underlying process 

Choosing a research design -

What is the purpose of the research?

Exploratory - find out what is happening, build theory. refine questions, not produce definitive answer - Descriptive - accurate profile of persons, events, situations, or pre-prepared questions to collect required information - Explanatory - seek answer to pre-specified questions. determine relationship - Population - quantitative methods - Sample - qualitative methods What is the research question? - should precede the design of the research study - the formulation of the question can imply aspects of the design required - decide on a research area - select a topic or issue within the research area - develop several alternative research questions - the questions should be articulated as a logical argument about observable/measurable things, events, people - Qualitative question: s hould not specify relationships or effects. what is it like? how do things occur? - Template for qualitative research question - How or what is the... - story of/account of - meaning of/perception of - theory that explains - issue of/experience of - Quantitative question: c  an form the basis for an hypothesis (predictions about the nature) - does the theory or model explain the relationship between IV and DV What are the pragmatic restraints? - Expertise: skills - Resources: funds, equipment, time - Access: reach the case, sample? - Control: c  onstrain certain factors to examine those of interest - Ethics: morally defined -

-

-



A good research question -

logical/simple expressed the question cannot allow for only one answer too broad a question or too many questions justifiable

Is it researchable?    

capable of being answered? resources? will the answer be of value? answerable within a moral framework is it discriminatory?...