Psych Paper 3 (& 2)

Ethical considerations (EC), `ECs in applying and reporting results, research methods etc. PAPER 2: research methods & eval - ABBREVIATIONS: IV = independent variable, DV = dependent variable, "Inv" = investigation, "co vars" = coefficient variables, "ptcp" = participant(s)

participant observations

• researcher participates as a member of the group they are observing

non-participant observations

• researcher does NOT participate in the group they are observing

focus-group interview

• interview with 6-10 participants

• rely on group process, and the interaction between participants/individuals

• allows for revelation of information not revealed between researcher + indiv ptcp

• resembles group dicussion

• interviewer acts as moderator

unstructured interviews

• interviewer has topics to cover, with a lot of freedom (has no direct order or precise Qs to ask)

• can include open & closed Qs

• interview evolves based on the interactions between interviewer + ptcp

semi-structured interviews

• has an interview schedule + topics to outline, but allows for elaboration/deviation

• resembles a conversation

• face-to-face discussion when researcher asking participants questions

• combination of open & closed ended questions

characteristics of case study

• in depth inv of an individual, small group or organisation

• multiple methods are used to gather data

• often a combo of quantiative + qualitative methods

• uses contextual analysis

• has longitudinal/retrospective design

• holistic approach to investigation/analysis

state the 5 sampling methods

1. random

2. convenience/opportunity

3. volunteer/self-selected

4. purposive

5. snowball

random sampling

• every member of the target pop has an equal chance of getting selected

• reduces sampling bias

• aims to obtain representative sample of target pop

convenience/opportunity sampling

• participants collected based on convenience/availability

• based on convenience for both researcher + participant

• may lead to sampling/researcher bias

volunteering/self-selected sampling

• participants approach researcher to participate

• usually participants root from marketing

purposive sampling

• participants selected based on relevant characteristics — invited to participate

• possible use of snowball simultaneously

• recruited through a range of methods (direct contact or referal)

snowball technique

an initial group of participants are selected, sample grows are those participants invite others.

• useful when studying “hidden pops” — pops that wish not to be found/hard to be found

(methodological/researcher) triangulation

the use of additional research methods OR researchers

why use triangulation

• increases credibility or validity (if similar results are found)

interviews (as a support for additional research methods)

follow-up interviews allow for the making-of-sense of the researcher’s findings in observations

ethical considerations for before reporting findings

• disclosing conflicts of interest

• fair reporting by the media

• informed consent

• anonymity

• right to withdraw

anonymity before reporting findings

participants names are removed from data — prevent participant’s possible embarrasment

debreifing before reporting findings

• asking ptcps follow-up questions ensuring the researchers findings are accurate

• ensuring participants are aware of how the study is to be reported

• ptcp can ask more questions about the study

fair reporting by media (before reporting)

anyone reporting results of a study have an ethical obligation to be truthful in how they report

informed consent before reporting findings

• informing ptcps of HOW the results are to be reported, in order for them to make an informed decision

disclosing conflicts of interest(s) (before) reporting

if the researcher has possible conflicts of interest/personal link in relation to what was investigated, it must be disclosed when reporting the results

verifying findings (when reporting)

• procedure of data should be shared w other researchers on controversial nature

ethical considerations for when applying findings

• avoid misinterpretation + misunderstanding

• ensure strengths AND limitations of research are understood

• reliability (has it been replicated)

• generalisability

• justice and equitability

reliability when applying findings

it could be considered unethical to apply findings of a study that has not been replicated

ways to better credibility

• triangulation

• larger sampling

• using controls

• replication/repetition of the study

define bias (+ researcher bias)

when existing thoughts/beliefs/ideas influence ones behaviour or thinking

researcher bias: when the factors above influence the results of a study

ways to avoid bias

• triangulation (method/researcher/data)

• different sampling methods (some are more prone to bias than others)

• using controls (double-blind design)

• replication

• reflexivity

reflexivity

the reseacher constantly reflecting on their role in the research and how it might be affecting them — this allows for the researcher to be more aware of their bias and thus try to reduce/prevent it

data triangulation (to avoid bias)

gathering data at more than one time to improve validity/credibilty of findings (increasing chances the findings being accurate reflection of reality and not being a random chance)

replication (to avoid bias)

• replicating a study and finding similar/same results increases credibilty/reliability of the study

define credibility

the extent to which the results of a study accurately represent(s) what was being studied/investigated

• qualitative research only

define validity

the accuracy of the methods of acquiring quantitative data (for what was being investigated)

• quantitative research only

factors affecting credibility [4]

• triangulation

• sampling

• controls

• replication

sampling to ensure credibility/validity

a larger sample = more representative of population = credibility/validity

controls to ensure credibility/validity

employing controls eliminate extraneous variables, confirming that the results are as a result of the manipulation of the var

possible controls include:

• counterbalance

• single- double- blind design(s)

• repeated measures

• control conditions

• random allocation of ptcps

  etc.

generalisability (or external validity)

the extent to which conclusions of a study can be applied/generalised to contexts beyond the study itself

transferability

the extent to which we can transfer the findings of a study to another context

• qualitative

things to consider doing when generalising/transferring the findings of a study

• how representational the sample is

• extent of ecological validity (bc of procedure)

• replicate

include S&L of RM, samplign tech etc