Research Methods Exam 2

Front 1

extraneous variable

Back 1

something that influences your study, even though it is not the thing you are interested in studying.

Front 2

four general types of extraneous variables

Back 2

participant variables, researcher variables, environmental variables, and measurement variables.

Front 3

Extraneous variables may vary

Back 3

systematically or unsystematically.

Front 4

A systematically changing extraneous variable

Back 4

one that changes with some kind of a pattern.

Front 5

An unsystematically changing extraneous variable changes

Back 5

at random, without a discernible pattern.

Front 6

A confounding variable

Back 6

a special type of an extraneous variable. A confounding variable systematically and simultaneously changes with target variables in a study. In other words, confounding variables are correlated with the variables a researcher is interested in studying.

Front 7

three general strategies for addressing extraneous or confounding variables

Back 7

The first is to eliminate them. The next way to deal with extraneous or confounding variables is to hold them constant. Suppose we are designing a study and think the age of the experimenter may influence the responses of the participants. We could keep this variable constant by using only experimenters who are a particular age. Finally, we can balance extraneous variables. This is the strategy behind the balancing and matched-groups assignment approaches mentioned in Lesson 3. Balancing involves planning how to distribute an extraneous variable across the study as a whole to control its effect. Balancing is often used in studies in which participants do a series of tasks, and researchers are concerned that the order of the tasks may influence their performance.

Front 8

Reliability

Back 8

the extent to which a measure or procedure dependably produces the same outcome.

Front 9

Validity

Back 9

the extent to which something measures what it is supposed to measure. Something must be reliable in order to be valid. However, just because something is reliable, it is not necessarily valid.

Front 10

Validity is not an inherent property of a measure. Rather, it is

Back 10

a description of a measure as it is shown in a particular sample.

Front 11

five types of validity

Back 11

face validity, construct validity, content validity, internal validity, and external validity. You don't need to know these last two for the progress evaluation or the exam.

Front 12

face validity

Back 12

measure has face validity, we are saying it looks like it measures what it is supposed to measure.

Front 13

construct validity

Back 13

the extent to which our instrument is measuring the underlying construct we mean for it to measure.

Front 14

content validity

Back 14

extent that a measure assesses all aspects of the construct it is purported to measure.

Front 15

the concepts of reliability and validity also apply to

Back 15

observed relationships between variables.

Front 16

two specific types of validity that apply to research designs

Back 16

internal and external.

Front 17

Internal validity

Back 17

extent to which our result is the result of the relationship between the variables we are actually interested in studying.

Front 18

external validity

Back 18

the extent to which we are able to generalize our findings beyond the participants in one study.

Front 19

There are several ways we can increase the external validity of our experiment

Back 19

first, we need to draw a representative sample. The more representative our sample is of the larger population, the more we are able to draw conclusions from our sample about the population from which it was drawn. Second, the more our findings are replicated (repeated) by other researchers and other experiments, the more likely we are to believe that we have uncovered a real phenomenon. Third, we can include as many "real life" elements as possible in our study.

Front 20

descriptive research

Back 20

examines existing relationships; no variables are manipulated or changed by the researchers.

Front 21

big advantage of descriptive studies

Back 21

they have good ecological validity.

Front 22

Disadvantages of descriptive studies

Back 22

First, researchers have little power to manipulate variables within a descriptive study. This means extraneous and confounding variables are more likely to crop up. Second, the results from descriptive studies do not provide strong evidence for cause and effect.

Front 23

There are three indices of central tendency

Back 23

mean, median, and mode.

Front 24

The mean (M) is

Back 24

the arithmetic average of all given scores and is computed by taking the sum of all scores (ΣX) divided by the number of scores (N): M = ΣX/N

Front 25

median is

Back 25

the score that divides the sample of scores in half.

Front 26

mode is

Back 26

the most frequently occurring score in the sample.

Front 27

several ways to index variability

Back 27

nicluding range, variance, and standard deviation.

Front 28

The range is

Back 28

simply the lowest and highest scores.

Front 29

Variance is computed by

Back 29

taking the squared sum of scores and dividing by the total number of scores. This value is subtracted from the sum of squared scores, and then this value is divided by the total number of scores. The formula for variance is as follows: S2=∑X2-(∑X)2/NN

Front 30

standard deviation

Back 30

square root of the variance.

Front 31

when we talk about the distribution of scores, we are really talking about

Back 31

variability.

Front 32

kurtosis

Back 32

the height or "peakness" of the distribution or the way that scores cluster around a particular point versus being spread out from it.

Front 33

platykurtic distribution

Back 33

the tails extend further, indicating more scores at the extremes.

Front 34

positively skewed

Back 34

the tail on the right (representing higher exam scores) is longer than the one on the left. In contrast, the distribution represented by the solid line is negatively skewed because the tail to the left (representing lower exam scores) is longer.

Front 35

four measurement scales

Back 35

nominal, ordinal, interval, and ratio. An acronym for these, in increasing order of information each provides, is NOIR.

Front 36

The most common type of correlation coefficient is

Back 36

the Pearson product-moment correlation coefficient, which is denoted as r.

Front 37

Pearson's r describes the

Back 37

strength of association between two continuously measured variables. Examples of continuously measured variables include reaction time, intelligence quotients, temperature, and exam scores, among many others

Front 38

two different subtypes of t-test

Back 38

the independent samples and the dependent samples t-tests, depending on how your data were selected.

Front 39

ANOVA

Back 39

"analysis of variance" and generates a statistic called F.

Front 40

If you find that the F value is_______, you can conclude there was some effect in your sample.

Back 40

significant

Front 41

We have been talking about determining whether a relationship is statistically significant. What does this mean?

Back 41

The first thing you should understand is that a finding of statistical significance does not mean a relationship is meaningful. Statistical significance is influenced by a variety of factors, including sample size, and it is possible to find statistical significance of a very small relationship with a large enough sample. This doesn't necessarily mean the relationship is meaningful. Alternatively, if a relationship does not reach statistical significance, this does not mean it is not meaningful. For example, suppose you find a very small positive relationship between a clinical intervention and a reduction in suicide rates. Even if the relationship does not reach statistical significance, we would all probably agree it is still a meaningful association.

Front 42

Statistical significance is

Back 42

a way of stating confidence that any relationship we have found did not occur by chance.

Front 43

the power of a test

Back 43

- o detect relationships, if they actually exist, is influenced by both sample size and effect size. A third factor that affects the power of a study is the level you choose for your p value, also denoted as α (alpha). The lower you set alpha (meaning the more confident you choose to be before accepting results as significant), the lower your power to detect the relationship will be.

Front 44

types of mistakes you can make

Back 44

First, you could reject the null hypothesis of no relationship (and thereby accept the research hypothesis), when in reality no such relationship exists in the population. This is called a Type I error. We can control the likelihood of making this type of error by setting α at a more restrictive range. On the other hand, if we set it too high, we might conclude that there is no relationship (i.e., accept the null hypothesis), when in fact there really does exist a relationship in the population. We would then have committed a Type II error. The probability of making a Type II error is denoted as β (beta). I found I was able to remember these by thinking that eager young researchers are likely to make their first error (Type I) by overanxiously concluding there is a relationship when there isn't.

Front 45

"Ethical Principles of Psychologists and Code of Conduct" (APA, 2002).

Back 45

This document establishes five general principles psychologists must adhere to when conducting research, providing therapy, or teaching. These five principles are listed in the box below. You do not need to memorize these five points, but I want

Front 46

• Beneficence and Nonmaleficence: Psychologists must help people and avoid harming them.

Back 46

asdf

Front 47

• Fidelity and Responsibility: Psychologists must act responsibly, professionally, and in ways that promote the best interests of others.

Back 47

adsf

Front 48

• Integrity: Psychologists must be honest, fair, and trustworthy.

Back 48

adsf

Front 49

• Justice: Psychologists must ensure that everyone benefits from psychology. They must also be alert for their own biases and the biases of the field. They must also be careful to practice only within their areas of competence.

Back 49

asdf

Front 50

• Respect for People's Rights and Dignity: Psychologists must recognize that everyone has rights to dignity and humane treatment. They must protect these rights, especially for those who may be disadvantaged.

Back 50

asdf

Front 51

psychologists must conduct their studies in a way that minimizes ____ and maximizes _____ to participants

Back 51

risks, benefits

Front 52

Informed consent generally includes

Back 52

information about the potential risks (psychological and physical), benefits (to oneself and others), and procedures involved in participation. It also should include a description of the purpose for the research, but this may be vague if too much knowledge of the study purposes could produce demand characteristics.

Front 53

Assent

Back 53

children and mentally retarded adults must also provide oral agreement to be in the study.

Front 54

Aftercare

Back 54

Psychologists have a responsibility to debrief participants and provide support after a study has ended. Debriefing involves giving participants more information about the study they participated in.

Front 55

How are ethics enforced

Back 55

First, students like you are taught about research ethics in classes like this. Next, any research project proposed by any student, professor, or other researcher must pass through an institutional review board (usually called an IRB) before research may begin. An IRB is a panel of people who carefully review all proposed research to ensure it is ethically sound. In particular, IRBs weigh the potential risks and benefits of each proposed study. IRBs exist at all universities and many other institutions. If an IRB rejects a research proposal, the research cannot be conducted. Furthermore, the American Psychological Association issues strict ethical guidelines its members must obey. These rules may be even stricter than the IRB, but they must be followed. If an APA member is found to be in violation of the APA Ethics Code, he or she can be removed from the APA. This is a grave consequence. It can jeopardize a psychologist's career, and he or she may be unable to find employment in psychology again. Finally, there are state and federal laws regulating the ways in which research can be conducted. Violations of these laws can lead to prosecution in the court system.

Front 56

demand characteristics

Back 56

function as confounding variables if they are not properly controlled.

Front 57

Demand characteristics are

Back 57

features in a study that somehow alter participants' responses or actions. Demand characteristics serve as subtle cues that tell participants how to behave.

Front 58

Three types of demand characteristics

Back 58

reactivity occurs when participants change their behavior simply because they are being watched, social desirability is the factor at play when participants decide to act in socially acceptable ways because they are in a study, experimenter expectancies are clues that researchers provide to participants about how they should behave.

Front 59

One of the best things researchers can do to minimize demand is

Back 59

to provide as few extraneous cues to participants as possible. It is important we do not provide participants with hints about how we hope they will behave. In order to minimize demand, it is often very important that participants are not aware which study condition they are in.