UBC PSYC 217 Final EXAM (2025) QUESTIONS AND (elaborated) WITH COMPLETE SOLUTIONS

Attrition

Another word for mortality

- People dropping out of studies

Informed Consent

Potential participants should be provided with all the information that might influence their decision to participate

In what types of research is multiple regression common

Personality, health, and longitudinal developmental research

Regression Equation

y = bX + a

Multiple Regression

Uses scores on one variable (the predictor) to predict changes in another variable (the criterion)

Squared correlation coefficient [r^2]

The amount of shared variance

- The % of variance in one variable accounted for by the second variable

r value indicates a [Medium] effect/covariation

0.30

r value indicates a [large] effect/covariation

>0.40

r value indicates a [small] effect/variation

0.10 - 0.20

Scatterplot

Each pair of scores from a correlation is plotted as a single point on a graph

Cohen's d = 1

The means are separated by 1 standard deviation

Cohen's d

Describes the magnitude of the effect of the IV on the DV

% of scores fall within 3 SD

99.7%

% of scores fall within 2 SD

95%

% of scores fall within 1 SD

68%

Variability

The spread in a distribution of scores

Mean as most useful measure of central tendency

It maximises the use of all data

- Can use in statistical analysis

Value of central tendency to use when you have [nominal or ordinal variables]

Mode

Value of central tendency to use when [outliers]

Median

Structural equation modelling

SEM examines models that specify a set of relationships among many variables

Yes/No is IV x PV true experimental design

No

- Participant variables are not randomly assigned or controlled

Statistical analysis to assess the unique contribution of each variable in a correlation

Multiple Regression

Discriminant validity

Is the measure not related to what it shouldn't relate to?

- The measure should be able to discriminate between the construct being measured and other unrelated constructs

Internal consistency reliability

Assesses how well a certain set of items relate to one another

- Is a participants score on this construct similar across items that are aimed to measure related aspects of the construct

Quality that single case experiments lack

Generalizability

Setting that true experiments generally harder to perform

In applied settings

Reactive measure show

What the person is like when he/she is aware of being observed

Type(s) if scale(s) allow you to make quantitative distinctions?

Internal and ratio

What an animal care committed is composed of

1 scientist with experience in animal use

1 institutional member who does not use animals

1 experienced vet and a community member

Multiple baseline: Across situations

Same behaviour is measured in different settings

Multiple baseline: Across Behaviours

Several different behaviours of a participant are measured over time

Multiple baseline: Across participants

The behaviour of several participants is measured over time

Ethical problem associated with reversal designs

You cannot ethically take people (medical patients) off treatment that is working

Reversal design (setup)

ABA

(A) Baseline > (B) Treatment > (A) Baseline

Evidence that the manipulation was effective in a single case experiment

A change in the participants' behaviour from the baseline to treatment periods

5 steps in performing a program evaluation

1 > Needs assessment

2 > Program theory assessment

3 > Process evaluation

4 > Outcome evaluation

5 > Efficacy assessment

Date of first major revision of TCPS

2010

Date of first TCPS published

1998

Set of ethical guidelines and policies Canadian researchers adhere to

Tri-council policy statement

Mixed Factorial Design

A mix of independent measures and repeated measures

Main flaw in cross-sectional designs

Differences may be due to cohort effects

You can't assume the next generation will look the same

Just research

Treat people fairly + equitably by distributing the benefits and burdens of the research

Concern for welfare in research

Attempt to minimize risk

And maximize benefit

Show respect for person in research

??

3 Basic ethical principles

1 Respect for persons

2 Concern for welfare

3 Justice

Ethical code that acted as a catalyst for modern ethical concern

Nuremberg Code (1949)

Consent forms content

Purpose of research

Procedures

Benefits + Risks

Any compensation

How confidentiality will be protected

Assurance of voluntary permission

Contact info

Higher order factorial

More than 2 IV's

Why REB's require that online studies are conducted using Canadian servers

US 'Patriot act' the US government can access records of any American internet servers

Risk-benefit analysis used to ensure

Concern for welfare

Naturalistic observations require an REB examination

No

2 Functions of debriefing

Ethical and educational

Popularity of elaborate deception since 1980's

Less Popular

When parents sign consent forms for their child

Assent

Multiple correlation

Symbolized as R

Provides the correlation between a combined set of predictor variables and a single criterion variable

Histograms

Uses bars to display a frequency distribution for a continuous variable

Pie charts are useful in representing

Nominal scale info

Variables name that are measured on interval and ratio scales

Continous variables

Interval or Ordinal scales - which is more sophisticated statistical analysis?

Interval

Fraud most often detected by

Colleagues or students working with the researcher

Two ways to commit fraud

1 Fabricating data sets

2 Collecting data from real P's but altering the #s to fit the hypothesis

The 3R's

3 things to do when working with animals

Replacement, reduction and refinement

Partial correlation

Provides a way of statistically controlling 3rd variables in correlational studies

Longitudinal Design

Observe one group of participants repeatedly over time

Give evidence for developmental change

Test-retest reliability coefficient value to determine the measure is a reliable one

0.80

Cronback's Alpha

Demonstrates internal consistency

Interrater reliability

The extent to which raters agree in their observations

Concurrent validity

Is the measure able to distinguish between theoretically relevant behaviour

Predictive validity

Using measure to predict some future behaviour

Content validity

Based on comparing the content of the definition of construct

Main problem with face validity

Very subjective

Face validity

Does it appear that the measure is measuring what it is supposed to measure

Construct validity

Construct operational definition

Are you hoping what you hope to measure

Reliability indicates...

The amount of error in a measurement

Convergent validity

The extent to which scores on the measure are related to scores on other measures of the same / similar constructs

Main problem with reactive measure

It doesn't tell us how the person would behave under natural circumstances

Two kinds of info yielded from a factorial design

Main effect

Interaction

Main effects

The effect that each IV, by itself, has on the DV

IV x PV design

How different types of people respond to the same manipulated variable

Simple main effect

Examines the mean differences at each level of one IV

Moderator variable

Influences the relationship between 2 other variables

Interactions

An interaction between IV's indicates that the effect of one IV is different at different levels of the other IV

Statistical variable used to address the reliability of a correlation

Pearson-product moment correlation coefficient

Reliability

The consistency/stability of a measure of behaviour

What needs to be guaranteed first when planning an experiment: Reliability or accuracy?

Reliability

Organization that oversees animal ethics

Canadian council on animal care

CCAC

Ratio scale

Has an absolute zero

Point that indicated the absence of a variable that is being measured

Interval scale

The difference between any 2 numbers on scale is equal in size

0 is arbitrary reference point

Ordinal scale

Allows us to rank order the levels of the variable being studied

Nominal scales

Has no numerical or quantitative properties

Is instead categories or groups

Cross-sectional design

Research design that examines people of different ages at a single point in time

IV within developmental research

Age

Efficacy assessment within program evaluation

Are the benefits of the program worth the costs

Structure of factorial design

# of levels of 1st IV x # of levels of 2nd IV

Factorial design

More than 1 IV and all levels of the IV are combined with all levels of the other IV's

Are complex designs still true experiment?

Yes

There is still random assignment

Required test for 3 or more variables

Variance test

What type of relationship does a complex design allow researchers to examine

A nonlinear design

Complex designs

Have multiple IV's and/or multiple levels

Most useful Quasi-experimental design

Nonequivalent control group pretest posttest design