psych 230: research methods exam 2

measurement that has a clear number; categorical/qualitative data

eye color, ice cream flavors, and nationality.

probability sampling: all population members have an equal chance to be selected as long as the population diversity is represented

non-probability sampling: all the populations don't have the same chance to be selected (leaving out ppl)

pros: Good to get a more accurate selection of people that can be applied to the real world.

cons: Might not be as readily available and harder for short time research

Pro: Easier to collect samples when don't have the necessary resources.

Con: The downside is that the experiment's results can’t be applied to the real world.

extent to which scores on measure represent the variable or construct they are intended to

1. face validity

2. content validity

3. criterion validity

4. discriminant validity

extent to which variable measures what it is supposed to measure

extent to which a test samples the behavior that we are interested in; extent to which a measure covers all aspects of the construct of interest

extent to which the measure is related to the outcome & other variables

Scores on the measures are not related to other measures that are theoretically different.

external variables that aren't controlled, decreased validity of results

This might be the age when looking at obesity and heart disease

1)Reaction 2) socially desirable 3)demand 4)expectancy

how people might react differently when they know they are being used for a study

subjects acting as they feel like is a desirable reaction

When subjects try to make sense of the experiment or get inside knowledge

Researchers may be biased in what they are looking at and what they expect to see

used for simple experiments, and generally just want to look and learn more about it

how accurate are first impressions

ranking order w/ no equal intervals

movie ratings, order of favorite snacks, rank stuff in order

extent of which the scores on a measure are consistent across time & multiple items on the same measure

redo experiment & compare results

1. reliability over time

2. reliability across items

3. reliability across researchers

How well your results should be through a specific time period

the extent to which items on a multiple item measure are consistent with each other. (questions should have some sort of connection to each other)

How consistent the observers are with their judgments.

how well your results are accurate and can research a conclusion on

standard error, standard deviation, and t-test.

Maturation Regression of the mean selection of subjects selection by maturation interactions mortality instrumentation testing history

a change over time that may change the results for the simple reason of growing

a person growing up will probably change answers when they are older or have more experiences.

Possible absence of subjects for whatever reason that can include death or not wanting to finish the study

experiment takes a long time, so some people are unable to continue the time commitment of the study

Change in measuring instrument or the use of instruments that might include proper calibration which can threaten validity

using older data might’ve been collected with different machinery and on different units.

Only taking the top 10% or bottom 10% to explain your results; the tendency that participants who receive extreme scores when tested, tend to have less extreme scores on subsequent retesting even in the absence of any treatment effects.

A person who scores an extreme value is less likely to do so if given the experiment again.

Any bias in selecting and assigning participants to groups that results in systematic differences between the participants in each group

non-random sample selection, cherrypicking certain people because it’s convenient or for other reasons

The treatment and no-treatment groups, although similar at one point, would have grown apart (developed differently) even if no treatment had been administered.

Boys and girls in 4th grade vs in 10th grade will be different.

When participants are repeatedly tested, changes in test scores may be more due to practice or knowledge about the test procedure gained from earlier experiences rather than any treatment effects

someone takes the same math test twice, gets a better score on the 2nd attempt.

Extraneous events occurring during the course of the experiment that may affect the participants’ responses on the dependent measure.

social upheaval may cause change in attitudes or behaviors in people.

comparing two pre-existing groups of subjects subjects to specific variables but do not alter such subjects. in medical research exposing subjects to diseases would be highly unethical Instead...take subjects that may already have the characteristics they are looking for and compare them

-In medical research exposing subjects to diseases would be highly unethical. -Instead take subjects that may already have the characteristics they are looking for and compare them

done by sending out questionnaires and relying on individuals to self-report their answers.

-relies a lot on the honesty of the individuals. -survey design is essential because it can affect the results.

meaning, there is order, and the difference between two values is meaningful because there are equal intervals between each score, but there is no true zero

temperature, pH, likert scale

does not involve the manipulation of variables.

It uses existing data or survey data that is collected. It is easier to do than experimental research and relies highly on participants' honesty. Correlation research does not measure causation, whereas experimental research can test for causation.

degree to which the changes in the dependent variable are due to the manipulation of the independent variable, and not by something else.

how much you can say that your results are due to your manipulations and not any other outside factors.

how the surrounding things affect how one might answer

include a survey with answers that look more off than others, and also the order of questions in the survey can impact how people answer

correlational research, archival research, single variable, etc.

There is no manipulation of variables, which makes it different from experimental research.

tool used to facilitate understanding of human behavior; observable complex traits, behavioral & internal processes w/ multiple operational definitions;

intelligence, self-esteem, and depression

includes name, natural order, equal intervals, and a true zero

height, the kelvin scale, and enzyme activity.

Qualitative research brings a deeper understanding than just quantitative data such as numbers. The numbers bring the data to the experiment, and it might be easier to conclude with the numbers.

involves the researcher changing that variable and viewing how the dependent variable changes in response to it. -important to see if there is any causation when changing the independent variable related to the dependent variable

ensures that the trend that the researchers are observing is only due to the manipulation of the independent variable, and not by any outside influence. -increases validity.

extent our results can be applied or generalized to the real world

Type 1: Ecological -generalizability of the results from one setting to another Type 2: Population -the ability to generalize from your sample to the population of interest

When choosing a nonexperimental, time, money, and resources are all factors

1)Naturalistic Observation 2)Participant Observation 3)Structured Observation

observing behaviors in the environment in which they occur.

researchers observe behaviors in the environments in which they occur, but researchers are active participants in the situation.

gathering quantitative data on a set of predetermined behaviors Ex: how many times do bowlers laugh or make an angry face after bowling?