EXAM 2 Flashcards
(124 cards)
1
Q
measures of central tendency
A
mean, median, and mode
2
Q
When might it be more appropriate to use a median as a measure of central tendency of a distribution rather than the mean?
A
When there are outliers
3
Q
X is…
A
a sample statistic
4
Q
u (mew) is…
A
the corresponding population parameter
5
Q
measures of variability
A
variance, standard deviation, range
6
Q
linear regression
A
finding the best-fitting line
7
Q
regression coefficients
A
slope and intersect
8
Q
y =
A
a + bX
9
Q
b =
A
slope
10
Q
a =
A
intercept
11
Q
positive slope
A
positive or direct relationship
12
Q
negative
A
negative or inverse relationship
13
Q
Pearson product moment correlation coefficient ( r )
A
-1 < r < 1
14
Q
2 pieces of info tells us if pos or neg
A
- The sign of r tells us whether there is a positive or
negative relationship. - r^2 tells us the proportion of variance in Y that is attributable to X (strength of the relationship)
15
Q
When r does not equal +/- 1.0:
A
- Only a portion of the variability in one variable is related to the variability of the other
- of one variable from the other will be imperfect
16
Q
r is a…
A
sample statistic
17
Q
P(RHO) is..
A
the corresponding population parameter for r
18
Q
Testing the significance of r..
A
- Is r sufficiently different from 0 for us to conclude that ρ ≠ 0?
- If the p-value is < level of significance (usually .05), conclude that ρ ≠ 0
- r(43) = -0.53, p = .002
19
Q
Multiple Correlation/Multiple Regression
A
Y = a + bX
Y = a + b1X1 + b2X2 + b3X3 + … + bnXn
- allows you to take more predictors into correlation
- typically linear
20
Q
nonlinear regression
A
data not in a straight line so formula is more complicated
21
Q
partial correlations and third variables
A
“partial put” the effects of the third variable
ex) pos. correlation between a city’s ice cream sales and drownings in city pools
- both are related to temperatures
22
Q
spurious relationship
A
when 2 variables are correlated but not causally related, but related by coincidence
23
Q
a correlation is..
A
just another piece of evidence that contributes to our understanding of something, not an end in itself
24
Q
measurement
A
assigning numbers or labels to observations to represent amounts or categories
25
measurement scale
set of possible numbers or labels that can be assigned to measured items
26
properties of measurement scales
1. magnitude (measurements can be arranged in order)
2. equal unit size (a change of 1 is the same size across the scale)
3. absolute zero (a score of 0 means there is none of what is being measured)
27
ratio
magnitude, equal unit size, absolute zero
ex) weight, salary
- in a ratio scale ratios have meaning ("twice as much money")
28
interval
magnitude, equal unit size
- no absolute zero
ex) temperature in degrees Fahrenheit
- 40 degrees is not twice as much as 20 degrees
- a particular difference is the same size anywhere on the scale
29
ordinal
magnitude
ex) rankings
0 difference between 1st and 2nd may be different that from 2nd to 3rd
- only tells you relative standing not distance between values
30
nominal
none of the three
- classification of items into groups that have no magnitude relationship
ex) nationality, gender, category names
- might involve numbers, but the numbers are just labels
31
Choosing a scale of measurement (nominal)
difference in quality
32
Choosing a scale of measurement (ordinal)
crude information about quanitity
33
Choosing a scale of measurement (interval)
also tells us how much scores differ
34
Choosing a scale of measurement (ratio)
tells us how much of the quality is/was present
35
reliability
repearability or consistency of a measurement (extent to which the measurement reflects the true value of what you are measuring)
36
2 components of measurement of reliability
1. true score ("indicate whether the following statement applies to you"
2. measurement error (poorly worded questions, unclear instructions, inconsistent measurement conditions, mistakes/inaccuracies by the measurer)
37
error adds?
variance
(The amount of measurement error is related to the
variance of the measurements)
38
test-retest reliability
Administer the same test at two different times
Compute correlation between 1st and 2nd scores
39
alternate forms reliability/Parallel forms reliability
Tests the equivalence between two different
forms of the same test (e.g., memory tests)
Administer both forms to many people
Compute correlation between scores on the two
tests
40
Internal consistency reliability
Split-half reliability
- Compute scores for two halves of the test
- Compute correlation between scores
on two halves of test
41
interrater/interobserver reliability
- Multiple observers rate many observations
- Compute correlation between ratings
42
Reliability is not the same as “______”
correctness
43
reliability vs. validity
reliability: Does the measurement instrument
give us the same answer every time? (e.g., thickness of notebook; ratings on a 7-point
scale for “happiness” items)
validity: Is the raw measurement a good operational definition of some construct? (e.g., Does the notebook thickness really reflect
“satisfaction with R.M. course”?; Does the score
on the questionnaire really reflect “happiness”?
44
construct
an idea developed to permit categorization and description of some directly observable behavior
psychological constructs are not directly observable
must be operationally defined
45
construct validity
is our measurement as operationally defined really measuring
what we think it is measuring?
46
Face validity
Is our theoretical construct meaningful? When we think we
are measuring the construct, are we really measuring that
construct, or are we really measuring some other, previously-
established construct?
47
Convergent validity
different measures of
the same construct give the same result
48
Discriminant validity
– measures of the two
constructs are not related, so they allow us to
discriminate between the two constructs (we
are measuring different things)
49
Evaluate convergent and discriminant validities with a
correlation _____:
matrix; Correlations on diagonal show reliability
(Remaining correlations are used to assess convergent
and discriminant validity)
50
Relationship between reliability and validity
- A measurement instrument may give the same
measurement every time (reliability), but it may be
incorrect (i.e., it may not reflect the true score of the
variable; there is a lack of validity).
- If a measurement instrument is not reliable (if it gives
a different measurement each time), it cannot be
providing the true value of the underlying variable you
are trying to measure (there is a lack of validity)
51
The most reasonable way to proceed is to assume that a
construct does not exist until ______ for it exists (test the
null hypothesis)
evidence
52
Criterion Validity:
The extent to which a measurement
instrument accurately predicts behavior in a particular
area
53
Predictive validity
give a test at some point in time,
compare score to future criterion behavior (GRE score and grad school performance; test of
coordination/attention and productivity on an assembly line)
54
Concurrent validity:
Give a test and then immediately
evaluate criterion behavior
Ex: pencil-and-paper test of navigation knowledge, then
immediately observe and rate person’s behavior in flight
55
Reactivity
subject’s behavior changes
because of awareness of observer
(allow subject to get used to observer and environment; conceal observer; allow for anonymous responding)
56
Nonreactive or unobtrusive measures
indirect observations of behavior
(Observe graffiti; count beer bottles in trash; Examine trash to assess food
consumption, recycling behavior, etc.)
57
Unobtrusive measures may or may not
involve intervention by the researcher
Evaluate fingerprint smudges on book pages to evaluate usage
Glue pages together and examine broken seals to determine usage (Friedman & Wilson, 1975
58
types of measures DV: behavioral
1. Record whether or not participant responds
2. Record the frequency of behaviors
3. Measure reaction time (latency)
4. Measure duration of response
5. Count number of errors or number correct3.
59
types of measures DV: physiological
§ Galvanic skin response - measure of general arousal and anxiety
§ Electromyogram - measure of muscle tension
§ Electroencephalogram - measure of electrical activity in the brain
§ Magnetic resonance imaging - produces images of the brain
§ PET scans of FMRI can show dynamic activity of brain
§ These are all Noninvasive techniques
§ Invasive measures - recording activities of individual neurons with microelectrodes
-Often, when using physiological measures you must infer psychological states from physiological state
60
types of measures DV: self-report
§ Participant might report beliefs, internal state, using a rating scale or yes/no response
§ Self report has potential reliability and validity problems
□ Participant might not be aware of true thought processes
□ Participant may not recall accurately
□ Participant may not behave in the future in the way that they predict
□ Participant may respond in a "socially desirable" way
61
types of measures: sensitivity of the dv
A rating scale can reveal smaller differences than a yes/no scale
Ceiling effect - task is so easy that everyone performs well
Floor effect - task is so difficult no one does well
62
Demand characteristics
- any cues or information about the experiment that might guide participant behavior
63
experimenter bias/expectancy effects
When experimenter expects a particular behavior and acts on away to cause that behavior to occur
64
Solutions to expectancy problems
train experimenters
print out instructions for participants so experimenters cant influence them in any way
automate data collection procedures
use experimenters who are "blind" to the IV manipulation
65
Double-blind experiment
neither experimenter nor participants know what conditions participants are assigned to
66
Research Proposal
Like introduction and method sections of a manuscript
Often peer-reviewed; provides opportunity to receive feedback
Gives organization to proposed experiments
67
Pilot Study
"trial run” for experiment
Shows potential for success
Reveals problems in design
68
Manipulation Check
Measures whether IV manipulation had intended
effect on participant
69
Reporting Results of Research
Professional meetings
Peer-reviewed journals
70
6 steps in constructing a questionnaire
1. Decide what information you are looking for
2. Choose a format
e.g., self-administered or used by an interviewer? (Consider using a pre-existing questionnaire)
3. write a first draft
4. reexamine and revise the questionnaire
5. pretest the questionnaire
6. edit the questionnaire and specify guidelines
71
survey formats: open ended (free response) questions
flexibility in responding
difficult to summarize; often requires a coding scheme
72
survey formats: closed questions
multiple choice, yes no, rating scale
easier to score and summarize
reduces expressiveness and spontaneity
73
Guidelines for writing survey items: simplicity
appointments are arranged expeditiously
vs.
there is excessive variability in the abilities of the doctors in this HMO
74
Guidelines for writing survey items: avoid double barreled items
When I call for an appointment, the appointment
clerk answers promptly and is courteous and
efficient
the furnishings in the waiting room are comfortable, and there is plenty of interesting reading material
75
Guidelines for writing survey items: avoid leading items
(items that suggest a particular response)
Most people accept the use of nurse-practitioners instead
of doctors for routine exams. What do you think?
Some people accept the use of nurse-practitioners,
some people oppose the use of nurse-practitioners, and
some people have no opinion one way or the other. What do you think?
What do you think about the use of nurse-practitioners
instead of doctors for routine exams?
76
Guidelines for writing survey items: avoid loaded items
(items that contain emotion-laden
terms or that suggest a socially desirable response)
I would be open-minded with respect to the use of radical new treatment plans.
I would be open-minded with respect to the use
of new treatment plans
77
Guidelines for writing survey items: avoid negative wording
the doctors do not order unnecessary tests, nor
do they fail to order necessary ones.
78
Guidelines for writing survey items: phrase items to detect "yea-saying" and "nay-saying"
A staff member promptly greeted me when I
arrived at the clinic.
My first impression of the clinic was that I was
being ignored
79
Guidelines for writing survey items: other
In multiple-choice items, make sure options are
mutually exclusive and exhaustive
Carefully define points on rating scales, particularly
end-points
Ask same type of question multiple times to check
reliability
80
Guidelines for writing survey items: use filter questions to avoid wasting respondents time
do you own a car? (If not, skip to question 47.)
81
Guidelines for writing survey items: Obtain demographic information for possible later use
(e.g. male/female, race, education level)
82
guidelines for writing survey items: Check for attention or random responding
“Fill in B for administrative purposes.”
83
Questionnaires
○ Inexpensive
○ Difficult to motivate respondents
○ Neither respondent nor researcher can seek clarification
○ Response rate can be low
84
Response bias:
some individuals chosen to respond to a survey systematically fail to do so
85
Interviews
○ More expensive to administer
○ Higher response rate, less chance of response bias
○ Respondent can ask questions, interviewer can ask for clarification of responses
○ Possibility of interviewer bias, when the interviewer inadvertently biases responses (…by reacting to responses; …by guiding responses)
86
Population
○ Large group you are interested in
87
Sample
Small part of the population
88
Statistics
○ Mean response, outcome
○ Numbers that describe the sample
89
Parameters
○ Mean of the population - population parameter
○Calculate statistics to get population parameters
90
Sampling
selecting a sample to draw conclusions about the population
91
Confidence intervals
EX) we have a 95% confidence interval that the true population mean falls within a given range
"46% of voters choose candidate x. the 95% confidence interval is + or - 5%"
= we are 95% confident that the true population value is between 41% and 51%
Identifies a numerical range that you are pretty sure contains the answer you are looking for
92
Representative sample
sample characteristics closely match population characteristics
93
Random sampling
avoids biasing
94
Self biased sample
Ex) questionnaire mailed to women's groups and placed them in magazines telling women where to write for a questionnaire
□ Not a random sample
95
Probability sampling techniques: simple random sampling
randomly select participants from the population of interest
96
Probability sampling techniques: stratified random sampling
□ Divide population into strata
□ Select a random sample (often of equal size) from each stratum
□ Guarantees that some of each group will be in sample
□ Might lead to overrepresentation of a group
97
Probability sampling techniques: cluster sampling
□ Useful in situations where you don’t have a list of all members of a population
□ Identify naturally occurring groups in the population
□ Randomly select clusters
□ Survey all participants in selected clusters
Easy and quick but people in the selected clusters may not be like those in other clusters
98
sampling techniques (nonprobability): Haphazard sampling (convenience sampling)
select whoever happens to come along
99
sampling techniques (nonprobability): purposive sampling
□ Similar to convenience sampling except that you choose participants that satisfy certain criteria
100
sampling techniques (nonprobability): quota sampling
□ Combination of stratified random sampling and haphazard sampling
□ Identify strata that are relevant to your topic
□ Select appropriate number from each strata ibn a haphazard way to yield same proportions as in population
101
sampling techniques (nonprobability): Snowball sampling
recruit participants and ask them to identify other potential participants
102
response rate
percentage of sample that responds (low response rate may indicate a biased sample)
103
observing behavior: quantitative approaches
choose a behavior of interest (helping behavior, eye contact, length of convo)
choose a measure of behavior (eye contact: yes/no; duration)
104
observing behavior: qualitative approaches
narrative record - record of behavior as it occurred (verbal written, video)
behaviors are then classified and organized
105
observing behavior: naturalistic observation
observation of behavior without any attempt to intervene
106
observing behavior: naturalistic observation (participation and concealment)
○ Participation - researcher becomes a member of the group that is under observation
problems: being a participant can affect objectivity/ observer might influence behaviors of others
○ Concealment - disguised vs. undisguised observation
addresses problems with reactivity (when observation changes subject's behavior)
§ In undisguised observation, use desensitization and habituation
107
observing behavior: naturalistic observation (advantages vs. disadvantages)
○ Advantages:
high degree of external validity
avoids some ethical problems
Useful for describing behavior and relationships between variables under natural circumstances (external validity)
○ Disadvantages:
No control by the researcher, therefore no inference of cause and effect (at best, correlational research)
○ Description and interpretation of data
Usually qualitative
108
observing behavior: systematic observation
study of one or more specific behaviors in particular setting
- not as broad in scope as naturalistic observation
- typically begins with narrative record
- then the narrative record in analyzed for occurrences of specific behaviors of interest
109
Sampling - an issue in systematic observation
typically all behaviors cannot be observed so how can you sample behavior to minimize bias and allow for generalizability (external validity)?
110
Time sampling (fixed or random)
- observe behavior only during certain time intervals, at fixed or randomly chosen intervals
111
Situation sampling
- sample behaviors in many different situations
112
coding system
(process of data reduction)
- interrater reliability/agreement: extent to which
the observations of different observers agree
- Raters are often trained extensively
- Carefully developed coding system leads
to precise descriptions
113
case study
detailed observation of a single person or situation, often to understand the development of some (usually uncommon) condition
- provide direction for future research, limited usefulness, cannot provide evidence to support one theory over another
114
testimonials
one person is telling you what happened in their particular situation
115
116
demonstrations
subliminal seduction (IN-CLASS)
117
naturalistic observation (case study)
type of case study that focuses on a particular situation
- has shortcomings of case study
Ex) Rosenhan: on being sane in insane places
□ If mental hospitals can distinguish between real and not real insane people
118
We are like case studies for ourselves:
cannot infer causation from only our own experiences
Ex: when we are sick, how can we be fooled into thinking some treatment cured us?
○ Placebo effect; cognitive dissonance
○ Spontaneous remission/disease ran its course/many diseases are cyclical
○ Hedged bets: seeking conventional treatment in addition to questionable treatment
○ The "worried well": those who seek treatment for illnesses they don’t have (Ear candling)
119
behavioral predictions are based on ______
probability
- there will always be exceptions and outliers
120
(case study) Archival research: statistical records
ex)
Instead of going to baseball games the researcher looked through records of baseball games where a batter was hit by a pitch
Looked at effects of heat as well as whether other players earlier in the game were hit by pitches
121
(case study) Archival research: written and mass communication records
letters, diaries, newspapers, magazines, tv shows
122
(case study) Archival research: content analysis
coding system
123
frequency distributions
indicates the # of ind. who receive each possible score on a variable
different types: pie chart (Nominal) bar graph (nominal & ordinal), frequency polygon (interval and ratio), and histogram (quantitative)
124
