Research methods in clinical psychology:An introduction for students and practitionersChris Barker, Nancy Pistrang, and Robert Elliott :Research methods in clinical psychology:An introduction for students and practitionersChris Barker, Nancy Pistrang, and Robert Elliott CHAPTER 10 The participants: sampling and ethics


Participants :Participants Sampling Ethics


Sampling :Sampling “Who?” who will you be studying? To whom can you apply the findings? Specifying and obtaining the sample quantitative and qualitative research Terminology “subjects” alternatives: participants, respondents, “co-researchers”


Sampling/ ctd. :Sampling/ ctd. Universe Target population Sample Intended sample Achieved sample


Generalising the results :Generalising the results (Generalisability = external validity) Sample population universe bias, plausibility error Sample size doesn’t determine generalisability


Sampling in practice :Sampling in practice Specify the target population Choose a sampling procedure Determine the sample size


The target population :The target population Inclusion and exclusion criteria Homogeneous sample (narrowly defined) reduces extraneous variability Heterogeneous sample (broadly defined) better generalisability Trade-off


Sampling procedures :Sampling procedures Goal: an unbiased sample every member of target population has equal chance of being selected Convenience sampling bias and under-representation of certain populations Eliminating bias: not always feasible gap between intended and achieved sample estimating bias


Determining sample size :Determining sample size Statistical power analysis depends on effect size, error rates Statistical power of a study = likelihood of it detecting an effect that is actually present


Power analysis :Power analysis Four interrelated parameters: Sample size, N Alpha (p-value in statistical tests): Type I error rate or false positive rate. Usually p<.05. The chance of a finding being wrong. ctd./


Power analysis/ ctd. :Power analysis/ ctd. Beta: Type II error rate or false negative rate (usually .20). The chance of missing something that is actually there Statistical power = 1 minus Beta Effect size: measures the strength of the relationship. Depends on the statistic used. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.


Alternative sampling approaches :Alternative sampling approaches Systematic replications Falsificationist approach look for counter examples Popper Snowballing/ networking Purposive sampling Theoretical sampling grounded theory approach


Ethics: overview :Ethics: overview Central principles: Informed consent Harms and benefits Privacy and confidentiality Ethics review committees Professional obligations


Ethics: general issues :Ethics: general issues Ethics is a methodological issue Procedural decisions have ethical implications Ethical problems arise from viewing participants as objects Ethical considerations require methodological trade-offs


Informed consent :Informed consent Full information Participants’ understanding needed Competence to provide consent Problems with deception Freedom of choice Voluntary decision (no coercion)


Informed consent form :Informed consent form Description of the study Explanation of its risks and benefits Use everyday language (no jargon) Offer to answer questions “You can withdraw your consent at any time …” Researcher’s and participant’s signature


Harms and benefits :Harms and benefits Minimisation of harm Direct harm: emotional distress, humiliation Indirect harm: withholding of benefit Harm to participants v. benefits to humanity


Privacy and confidentiality :Privacy and confidentiality Privacy: right to not provide information to researcher Confidentiality: right to withhold information from third parties Protecting confidentiality: anonymity research codes audio/videotapes Limits to confidentiality


Cost-benefit analysis :Cost-benefit analysis risks versus benefits greater risks of new procedures benefits to participants versus benefits to others greater potential risks require stronger safeguards


Ethics: conclusions :Ethics: conclusions Ethics review committees Professional obligations to be familiar with code of conduct/ethical principles to respond to participants who need professional services Difficult choices: seek advice