TCPS first standard canadian ethics code for research Nuremberg Code comment im research principles in the TCPS 2 Justice: Jain mess to participants Respect for persons: autonomy Com team for welland: benefit - risk analysis C benefits - e . g. monetary ; cost of conducting the study ninks- physical harm; stress (noteguards/debrief sessions); loss of confidentiality (data usage; questions need to be confidential and responses anonymous ) - careful when using personal data R: informed concent; concent form needs info that affects choice to participate ; cony language; coencion ( threat to B; too great compensation); assent (minor); secondary use of data ( threat ); too much impo is not necessary ; deception ( unknowing participant ; create false purpose / story ; une when informed consent is bias ); debriefing- needed is deception; explain why; Windsor Deception Check list : see if they should use role-play ( problem - demand change . ) simulation or homest studies J: all decisions need justification; equal benefits and burdens REB Research ethics board; studies need approval ( changes too ) no need if use public observation on impol anonymous on public)- this is exempt research minimal risk research-rink im research = daily-life mink Amimo better for long-time Benconch : study/condition control -> Canadian Council Animal Care 3RD: replacement (alternatives); reduction ( minimize number of animals used); refinement (minimize pain / distress) needs supervision / training / cane
(PA) ethics code in psychology ; consistent with TCPS 2; scientific misconduct/ false evidence); proud-making up on altering data + detected on peer review ; fail replication ; do it for leon of failure and pressure; intentional errors; homest science: disclosure ( record of study); preregistra- tion ( public statement); open data and materials (online ) Reliability parme result when you measure some thing true score: person's actual level of the variable of interest measurement enronDany that contributes to score that in mot based on true score careful training: maximize true I number of items: I reliability Pearson: correlation coefficient(n); - 1.00 ( sme variable increases, other decreases); +1.001 both increase); 0.00 ( variables are um related ) types of reliability Test-retest give some measure twice ; une pon developing scale Internal degree that's items are consistency : consistent when measure some variable (nombach's average correlation of each item in a me asune ; use to brief version of a measure split-half reliability: determine degree that all items in scale are related to one another-split scoren and calculate correlation Inten-nater degree to which naters agree with each other ; Cohen's kappa (indicator) Validity: mcanvas what it claims to? internal causal claim/ well-designed construct measure measures the intended phenomenon indicators of construct validity Face content of measure appears to reflect construct measured Content content of measure captures all necessary aspects of construct, nothing mor