PSYC 2305 Final Exam Review

Scientific knowledge is more trustworthy when it is:

based on empirical evidence and peer-reviewed research.

Scientific knowledge is more trustworthy when it is

• intersubjectively testable

• reliable

• denifite/precise

• coherent

• comprehesive

• open to change

Scientific Hallmarks

use of empirical methods, systematic observations, public knowlege (peer review)

Standard Form

What is the effect of the IV on the DV?

Con

Construct

concept/idea we can’t directly observe or measure

• some postulated attribute of people, assumed to be reflected in test performance

the scope of generalization

to answer a big question, we must make it small but making it small does not necessarily answer the big question

operational definiton

public, objective description of what the investigator did to study the variable or the set of procedures used when you measure or manipulate the variable

What type of study is lowest in control?

Naturalistic Observation

Case Studies

clinican studying patient

Archival studies

a source of hypothesis for future testing.

clinician studying her own records

Retrospective (ex post facto)

a category of research design in which the investigation starts after the fact has occurred without interference from the researcher.

Intervention studies

high in control, avoids confounding.

• has external validity & generalizes to others

• is accomplishd by proper sampling (random sample)

true experiments

involve manipulation of IV and random assignment

• high in control

well-controlled experiment

random assignment, internal validity, protects against confounding variables

temporal precedence

the cause precedes the effect

covariation of cause and effect

when the cause IS present,the effect also occurs.

when the cause is absent, the effect does not occur.

the fundamental threat

the confounding variable

confounding variable

a variablethat systematically varies with the IV is confounded with DV.

• cannot conclude the IV caused the change in the DV as the CV might have caused it

Intrinsic confounders

participant traits such as age, gender, race

extrinsic confounders

situational factors such as practive or relevant experience

possible confounders

• intrinsic traits

• extrinsic traits

• self-selection

• experimenter bias

• pygmalion effect

Pygmalion effect (Rosenthal effect)

The phenomenon where higher expectations lead to an increase in performance. It illustrates how expectations from teachers or supervisors can influence the achievement of individuals.

paradox of the confounding variable

there is more to a CV than covariation. also need a plausible between the EV and DV

controlling confounds

• random assignment & matched pairs

• double-blind procedures

• controlling order effects (counterbalancing)

Factor

another term for independent variable

levels

variations of the IV

between-subjects design

different participants in each condition

within-subjects design

same participants in all conditions (risk: order effects)

mixed design

combines both between-subjects and within-subjects experimental designs

placebo effect

a phenomenon where participants experience a perceived improvement in condition due to their expectations, rather than the actual treatment itself.

• similar to the experimenter expectation effect

control refers to the control over…

the research situation, not the participants in it

significance

does the IV have an effect on the DV?

-shown if at least one box in the matrix has a different number

Multifactor Standard Form Question

What are the effects of Factor I and Factor II on the DV?

first P of PETMOP

participants

E of PETMOP

experimenters

T of PETMOP

treatment

M of PETMOP

materialsthe resources needed for the experiment, including tools, measurements, and stimuli.

second p of PETMOP

placebo

an ________ exists when the effects of one factor depends upon the level of another factor

interaction

what does an interaction look like on a graph?

non parallel lines

2×2 design

2 factors + 2 levels

2 main effects + 1 interaction to evaluate

3×2 design

3 levels in factor I

2 levels in factor II

6 cellsto analyze, with potential for 2 main effects and 1 interaction.

inherently flawed design

single factor, two-level, within subject design because the levels of the IV will confound with the order of testing

a ____________________ exists when the effect of one factor depends upon the levels of both of the other two factors

three-way interaction

three-way interaction composition

3 main effects, 3 two-way interactions, 7 outcomes

how many outcomes does a four factor study have?

14 outcomes

x axis of frequency distribution

score/IV

y axis of frequency distribution

frequency/count of scores

measures of central tendency

mean, median, mode

when should you use the median instead of the mean?

if there is an extreme high value or extreme low value that will skew data

measures of variability

range, variance, standard deviation

range

difference between high and low

standard deviation

the square root of the average of squared deviations from the mean

descriptive z score formula

= (X - μ) / σ

x = data point

μ = population mean

σ = standard deviation

what does descriptive z score do?

describes how many standards deviations a specific value is away from the population mean in standard deviation units

what is the use of the descriptive z score?

used to describe the position of a single data point within a distribution

what percentage is one standard deviation (SD)?

68%

variance

the sum of the average of squared deviation from the mean

what percentage is two standard deviations (SD)?

95%

what percentage is three standard deviation (SD)?

99.7%

correlation coefficient, r

represents degree of relationship between variables, ranging from -1 to +1.

positive correlation

indicates that as one variable increases, the other variable also increases, and vice versa.

direct relationship

negative correlation

indicates that as one variable increases, the other variable decreases. This reflects an inverse relationship between the two variables.

what does restricting the range do?

lowers correlation value

validity

the correlation of a test with an external criterion

inference leads to the…

possibility of error

decision rule is…

set before starting

alpha level, a

state how willing you are to make an error

the tolerance to making a mistake and going from skepticism to slight acceptance

if less than a…

it is very rare

rare =

5% of the time or less, not less than a

parameter

value fo the population

statistc

value of the sample

null hypothesis testing

1. Research hypothesis

2. Null hypothesis

3. Determine alpha level.

4. Obtain a random sample from population; compute mean and standard deviation.

5. Compute unit size (SE) for the sampling distribution.

6. Determine how many units X is away from the mean of the sampling distribution, n. compute z-score associated with sample mean.

7. Determine what percentile it corresponds to (using the table)

8. Decide if that percentile is smaller than a

type I error

false positive

• rejecting a true null hypothesis

  • indicates there is an effect or difference when none actually exists.

type II error

false negative

• failing to reject a false null hypothesis

  • occurs when a test fails to detect an effect or difference that is present.

sampling distribution

distribution of the difference of scores between means

• normal shape

• equal to the true mean of a population

standard error

equal to the standard deviation of the sampling distribution divided by the square root of the sample size.

• the standard deviation of the sampling distribution

inferential z score formula

= (x - x̄) / (σ)

x̄ = sample mean

σ = population standard deviation

when should we use inferential z score?

we want to know where an individual sample mean lies within a distribution of sample means

what does the inferential z score do?

describes the position of a single data point within a distribution

what is inferential z score used for?

used to determine statistical signifcance

to make inferences about a population based on a sample

true score

actual value

how do you calculate the observed score?

true score + error component

what is the observed score?

the value we measure

what are the 3 ethical principles based on the Belmont Report?

beneficence, autonomy, and justice.

benefience

maximize benefits, minimize harmful effects to participants

autonomy

respect for participants’ decisions (principle of informed consent)

justice

fair distribution of research burdens and benefits

informed consent

• purpose of the research

• procedures used

• risks and benefits

• confidentiality

• assurance of voluntary participation

insititutional review boards (IRB) is a…

federal requirement (practical mechanism)

how many IRBs does the University of Houston have?

3

one of the cornerstone principles of science is…

replication