PSY 230 Quiz 2
Non-directional
Different Short term mating orientation will differ between men and women (Doesn't say WHAT the difference is)
directional
Larger/smaller, higher/lower, positive/negative association (Adding a specific direction) Men will have higher STMO score compared to women
Content validity
- Do the items measure all aspects of what we want to measure? • Example: depression - affect, behavior, physical • When you want to know whether a sample of items truly reflects an entire universe of items on a certain topic • How to Establish . . . - Content expert - Do items represent all possible items? - Course or Class: How well does the number of items reflect what was taught?
Criterion validity
- Does the measure correlate (or not correlate) with theoretically relevant items? • Concurrent criterion validity: • Predictive criterion validity: extent to which a measure is related to an outcome -• When you want to know if test scores are systematically related to other criteria • Concurrent criterion validity: How well a test outcome is consistent with a criterion that exists in the present - Personality measures with theoretically-related measures • Predictive criterion validity: How consistent a test outcome is with a criterion that occurs in the future - A measure of CVD risk with hard measures of CVD outcomes
degree of deviation
- Make scores interpretable (where are you relative to the mean?) • Z-scores measure the _______________ from the mean in units of standard deviation (+/-) • Usually ranges from -3 to +3 - Mean = 0 - SD = 1
reliability coefficients
- The reliability coefficients to be positive and not to be negative - Reliability coefficients that are as large as possible (between 0.00 and +1.00) • ~.75
sources of error
- testing situations or conditions that impact the obtained score rather than the qualities of the trait being measured. - Example: Whether the temperature of the room impacts the test score
normal distribution
- unimodal, - bell-shaped - symmetric around the mean
z score
1. transform raw scores into scores that have more information (exact location of the X value in the distribution) 2. standardize distributions (allows comparison to other distributions that have also been transformed into z-scores)
sign and number
2 components of z score
null/research
2 major types of hypothesis
z-scores 0 to +1
34.13% of scores
z-scores 0 to +2
47.72% of scores
z-scores -1 to +1
68.3% of scores
z-scores -3 to +3
99.7% of scores
directional research hypothesis
A _______ is a hypothesis that reflects the difference between groups and also specifies the direction of the difference.
testable
A hypothesis is a ________ Statement: o Translates the problem or research question into a form that can be tested Ex of research statement: how does experiencing weight stigma affect eating?
z-score
A standard score that is the result of dividing: A. the amount that a raw score differs from the mean of the distribution by B. the standard deviation
unsystematic
Stuff you can't really control for (Noise) - Randomize, Counterbalance
z score
The _______ is the most commonly used standard score.
true score
The actual or measured score is called the _______.
test-retest reliability
The correlation between scores from Time 1 and Time 2 is called _______.
asymptotic
The fact that the tails of a normal distribution never touch the horizontal axis relates to the following property: ______.
sample
The group you actually collect data from for your study is known as the _______.
independent variable
The more violent television programs an individual watches, the more likely the individual develops aggressive behavior. Watching violent television is ______.
mean and standard deviation
The normal curve is... - defined by 2 PARAMETERS:
dependent variable
The outcome variable in an analysis is also called the _______.
True score
The true, 100%-accurate reflection of what you really know - Rarely are the observed and true scores the same.
14%
Under the normal curve, approximately what percentage of scores falls between -1 and -2 standard deviations below the mean?
z score
What type of standard score has M = 0 and SD = 1?
observed score
What you actually get on a test (e.g., 89 or 65)
Parallel-forms reliability
When you want to know if several different forms of a test are reliable or equivalent - Example: Split items up that purport to measure the same thing and correlate
Internal-consistency reliability
When you want to know if the items on a test assess one—and only one—dimension - Example: do all these items go together?
Test-retest reliability
When you want to know whether a test is reliable over time - Example: A personality test taken and then taken again a month later
Interrater reliability
When you want to know whether there is consistency in the rating of some outcome - Do these raters agree
There is a positive relationship between a high-fat diet and weight gain.
Which of the following is an example of a directional hypothesis?
they are negative
Which of the following is true of z scores that fall below the mean?
predictive validity
Which of the following validities is concerned with monitoring estimates of future performance?
Construct validity
_______ is based on the judgment of how well a test reflects an underlying idea.
content validity
_______ is established by consultation with an expert on the topic focused upon by your instrument.
Research hypotheses
can be directional and nondirectional
Platykurtic
flat distribution
Kurtosis
how peaked and pointy the distribution is
independent variable
is the treatment.
reliability
means that the measure is consistent in measuring what it was designed to measure. It is whether a test—or whatever you use as a measurement tool—measures something consistently
Validity
means that the test, scale, or instrument measures what it is supposed to.
null
nothing, no difference, no inequalities between groups
true
null hypothesis is accepted as ___ without knowing more information
z score
number of standard deviations that a raw score is above/below the distribution mean
raw score
original, unchanged score that is direct result of measurement (X) without context, raw score doesn't tell you much e.g score of X = 76
Null hypothesis
refer to populations and are written in Greek Symbols
Research hypothesis
refer to samples and are written in Roman symbols
standard deviation
the amount one would intuitively expect that a random observation would differ from the mean
Construct validity
the degree to which it measures - Do all these items go together? • How well a test reflects an underlying idea or construct - intelligence or aggression • Most interesting . . . most difficult source of validity to establish • Want your construct to correlate with related behaviors and not to correlate with behaviors that are not related
Population
the large group to which you would like to generalize your findings
sampling error
the measure of how well a sample represents the population o Deviation away from the population parameters o Example: interested in pregnant latino mothers; you would not take a sample of males
sample
the smaller, representative group of the population that is used to do the research
Leptokurtic
very peaked
dependent variable
what the researcher looks at to see whether any change has occurred as a function of the treatment.
valid
• A ______ test is one that measures what it is supposed to measure.
distribution
• A ______________ is the summary of how likely (probable) a value or range of values may be. The area under the curve indicates this likelihood.
reliable
• A test can be _______ but not valid.
valid
• A test cannot be ______ unless it is reliable because if a test does what it is supposed to, then it has to do it consistently to work.
Hypothesis
• An educated guess - informed by past research or theory
significance
• How unlikely an event (or association) might be - P < .05 - z-score of +1.96 (P-values Smaller is better!!) - When your probability falls below 5%, we assume that the event may have some reason for occurring (not by chance)
Establishing reality
• Increase the number of items. • Delete unclear items. • Make the test neither too easy nor too hard. • Reduce the effects of external distractions.
normal curve
• Provides a visual representation of a distribution of scores • Mean, median, and mode are equal to one another • Tails are asymptotic (get closer to horizontal axis but never touch it)
Directional Research Hypothesis:
• Reflect a difference, and the direction is specified • Use a ONE tailed test
non-directional research hypothesis
• Reflect a difference, but the difference is not specified • Use a TWO-tailed test (Use this because the direction is not specified) • H1 : X1 ≠ X2 • States that they are not equal to each other
research hypothesis
• Statement that there is a relationship between two variables • Statements of inequality • There are two types of research hypothesis
Null hypothesis
• Statements that contain two or more things that are equal (unrelated) to each other
Probability
• The areas of the curve that are covered by different z-scores also represent the __________ of a certain score occurring.
z score
• They tell you where your score was relative to scores drawn from a normal distribution.
Probability
• To determine the likelihood that an effect/difference was generated by random chance (the Null Hypothesis).
reject
• We cannot prove the research hypothesis BUT we can _____ the null
Probability
• We reject the Null Hypothesis based on _______ - This effect was likely not caused by chance - We can infer that the effect (relationship) exists but we cannot prove it • Normal Distribution
nature
• What makes the normal distribution special is that many processes in _____, when randomly sampled, produce observations that are normally distributed.
hypothesis and Z scores
• When we put together_____________ you will be able to estimate how likely certain outcomes (specified in the research hypothesis) are.
likelihood
• With the mean and standard deviation, you can predict the ______ of ANY real number drawn from this distribution.
good hypothesis
• be stated in declarative form • posit a relationship between variables • reflect a theory or a body of literature on which they are based • be brief and to the point, and • be testable.
systematic
Deliberate like a manipulation (Signal)
observed - true
Error =
Skewness
How lopsided or not symmetrical the distribution is
mean and standard deviation
If you want to calculate a z score for a test where your raw score was 24, what other information must you know?
construct validity
If you want to know that a test measures some underlying psychological construct, what type of validity evidence would you want to collect?
test-retest reliability
If you want to know whether a test is reliable over time, what type of reliability do you use?
dependent variable
If you wanted to examine the impact of fast-food consumption on weight gain, your measurement of weight gain would be the _______.
observed score
In terms of the reliability of test scores, there are multiple elements to each person's score. The score that is actually recorded is the _______.
normal distribution
Large samples approximate a
Variability
Spread across numbers