Research methods notes FINAL.docx

14 Pages
Unlock Document

Family Relations and Human Development
FRHD 3070
Georgevander Merwe

Research methods notes Chapter 12 – qualitative interviewing Qualitative interviews - Participants are active participants - Informal and non directive o Mutual sharing of experiences between interviewer and client o Focus is on clients perspectives and experiences - Developing rapport is essential Length of qualitative interviewers - Not necessarily a clear beginning or end o Can be picked up at a later time Structure of qualitative interviews - Questions in no particular order tailored to the interviewee - Interviewer asks open questions and encourages elaboration - Inductive approach to theorizing, no testing hypothesis Differences between qualitative interview and a friendly conversation - Obtain consent from interviewee - Interview has a purpose - Interviewer asks most of the questions - Repetition is used more often for clarification - Formal ending to interview Sampling for qualitative interviews - Typically select through non-probability sampling - Snowball and purposive sampling used for recruiting interviewees o Purposive  Find a diverse group of people to study with a purpose for selection o Snowball  Select 1 or 2 participants and get them to refer more participants and so on - Used to study hidden populations with no sampling frame How many people to interview - Continue interviewing until theoretical saturation is achieved o Often dictated by the time and resources available Incentives - A general term for the gift given to research participants o Often in the form of cash - Use incentive appropriate to target sample o 10$ fast food coupons to street yout Interview sites - Often take place in interviewees home where they are comfortable o If no privacy or distraction, interview moved to another setting (resteraunt) Interview meaning - Shaped by gestalt o Body language, relationship between interviewer and interviewee, and the context  Interviewer notes non-verbal symbols with meaning (e.g. shrug or head nod) Recording and transcribing - Selective transcription o Interviewer only notes relevant and important information  Fully transcribed interviews are most dependable Informants - A member with whom the researcher has a relationship, and reports on the aspects of the research setting - 4 keys to a good informant o Informant is very familiar with the culture  Engages in routines of the culture without thinking about them o Currently involved in the culture  Current members are most useful for getting correct information o Person can spend time with the researcher  Must be available for full duration of the interview o Nonanalytical individuals make better informants  Familiar with and uses native folk theory or pragmatic common sense Kvale’s question types - 9 different types o Introducing  What comes to your mind when you think of…? o Follow up  Get additional despcription of topics just discussed o Probing question  Expand on incomplete points that interviewee has made o Specifying questions  Get more detailed descriptions about specific aspects o Direct questions  Towards the end of the interview to address topics not yet discussed o Indirect questions  Get a sense of how the interviewee thinks or feels about other people o Structuring questions  Keep the interview on track or keep it moving alonog o Interpreting questions  Interpreting what the interviewee is saying as correctly as possible o Silence  Technique used b interviewers to get interviewees to continue speaking Interview guides - List of questions researcher wants to cover o Not in order it is just a guide - Used towards end of interview to make sure all topics are covered Advantages of qualitative interviews - See the world from perspective of interviewee o Interviewee is free to speak using whatever language and words they wish - Lots of rich detailed data o Fewer interviewees but far more detailed data - New theories can emerge o Because it is inductive approach  New theories developed using grounded theory approach o Inductive approach  Theory comes from data collected - Development of new avenues of research o Create theories that feed into future research  Can also study things that have barely been studied Limitations of qualitative interviews - Too much data can be time consuming and overwhelming o Even though sample sizes are usually small Problems with validity and reliability - Small sample sizes make it difficult to relate to wider population - Lincoln and Guba suggest that qualitative research should be evaluated according to trustworthiness o Make it trustworthy by allowing other esearchers to renalazye it Focus groups (group interviews) - Group of people informally interviewed in a group discussion setting - Popular way for market researchers to test their product - Topics might include o Public attitudes (workplace equality) o Personal behaviour (avoiding STIs) o A new product o A political candidtate The focus group procedure - 6-12 people in a room with a moderator o Moderator = trained facilitator who guides the focus of the discussion - Most last about 90 mins - Held in rooms equipped with audio and video recording so sessions can be transcribed Composition of focus groups - Determined by research questions - Members should be homogenous and not friends or relatives Number of groups in a focus group study - Typical study researcher uses 4-6 groups o Time and money often indicate number of groups - As many focus groups as possible to reach theoretical saturation Strengths of focus groups - A lot of content and depth in answers - Easily attain info from illiterate individuals or children - Participants can speak in their own words Chapter 9 – experimental research Experiments - Use the positivist approach o Begin with a hypothesis o Modify something o Compare the outcome with and without the modification - Views causation o Best way to see cause and effect Limitations - Limited to studies where researchers can manipulate conditions and intervene - Best for micro level focus o Have limited generalizability and cannot address questions across decades Randomization - Random assignment o A method of assigning cases to groups for the purpose of comparison  Increases the confidence that groups do not initially differ in systematic ways o Begin with a selection of cases and put them into groups randomly (flipping coin - Random o Each case has an equal chance of being selected True experiment - Parts of the experiment o Treatment/independent variable  Independent variable o Dependant variable o Pretest  Measure of dependent variable prior to treatment o Post-test  Measure of dependant variable after treatment o Experimental group  Group that receives the treatment o Control group  Group that does not receive treatment o Random assignment Control in experiments - Control all aspects of the experiment to isolate the effects of the independent variable o Eliminate alternative explanations - Control the experiment using deception o Interviewer intentionally misleads subjects of the true nature of the experiment o May involve confederates or stooges  Pretend to be subject or bystander even though working with researcher Types of design - Experimental design o Arranging the parts of the experiment and putting them together - Classical experimental design o Design that has random assignment, control group, experimental group, pretest and postest for each group - Pre experimental designs o Lack random assignment, much weaker than classical o Make inferring a casual relationship more difficult o One shot case study design  Only one group, a treatment and a post test  No random assignment o One group pretest – posttest design  One group, pretest, treatment, post-test,  No control group or random assignment o Static group comparison  Two groups, no random assignment and only a posttest  Limitation • Any posttest differences could be due to difference in group rather than treatment Quasiexperimental designs - Stronger than pre-experimental designs - Variations of the classical experimental designs (quasi=variation) o Used in special situations or when the experimenter has little control over the independent variable - Two group – post-test only design o Identical to static group comparisons with one exception  Groups are randomly assigned o All parts of classical design except pretest - Interrupted time series design o One group, makes multiple pretest measures before and after treatment o Measures dependent variable periodically across many time points  Treatment occurs in the midst - Equivalent time series design o One group design that extends over a period of time o One group, pretest, treatment, posttest, treatment, posttest, treatment etc. - Latin square designs o How several treatments given in different orders affect dependent variable - Solomon 4 group design o Randomly assigned, 2 control groups, 2 experimental groups,  Only 1 control and 1 experimental receive pre test  All 4 have post test o Addresses issue of pretest effects - Factorial designs o Considers the impact of several independent variables simultaneously o Interaction effect  The effect of two independent variables that operate simultaniouslu Design notation - Name of a symbol system used to discuss parts of the experiments and make diagrams of them - 0 = observation - X= treatment - R = random assignment Internal and external validity - Internal validity o Ability to eliminate alternate explanations - Threats to internal validity o Selection bias  Groups are not equivalent at the beginning of the experiment  Problem without random assignment o History effects  Something that occurs and effects the dependent variable that is unplanned and out of experimenter control  Talking about attitudes towards guns, family member killed by gun during experiment o Maturation  Due to natural process of growth, boredom, and so on that occur during an experiment  Students become bored and sleepy and score lower o Testing effect  Pretest effects posttest  More than treatment alone effect results  Subject remembers pretest questions and gets them right on post-test o Instrumentation  Dependent measures variables change during the experiment  Springs on scale weaken during weighing experiment o Mortality  Subjects failing to participate through entire study  Start with 50 subjects and end with 30 o Statistical regression  Random errors move group results towards the average  If you begin at an extreme they can only go towards the average o Diffusion of treatment and contamination  Different groups discuss and learn about eachothers treatment o Experimenter expectancy  Researcher indirectly communicates desired findings to the subjects Mediator vs. moderator - Mediator o Things in between intervention and outcome - Moderator o Explains relationships, looking at moderator can strengthen, weaken, or neither the relationships between independent and dependent variable Double blind experiment - Neither subjects nor person who deals with subjects knows the specifics of the study - Placebo o False treatment or one that has no effect o Subject mistakes for a true treatment Ext
More Less

Related notes for FRHD 3070

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.