Busy. Please wait.
or

show password
Forgot Password?

Don't have an account?  Sign up 
or

Username is available taken
show password

why


Make sure to remember your password. If you forget it there is no way for StudyStack to send you a reset link. You would need to create a new account.
We do not share your email address with others. It is only used to allow you to reset your password. For details read our Privacy Policy and Terms of Service.


Already a StudyStack user? Log In

Reset Password
Enter the associated with your account, and we'll email you a link to reset your password.
Don't know
Know
remaining cards
Save
0:01
To flip the current card, click it or press the Spacebar key.  To move the current card to one of the three colored boxes, click on the box.  You may also press the UP ARROW key to move the card to the "Know" box, the DOWN ARROW key to move the card to the "Don't know" box, or the RIGHT ARROW key to move the card to the Remaining box.  You may also click on the card displayed in any of the three boxes to bring that card back to the center.

Pass complete!

"Know" box contains:
Time elapsed:
Retries:
restart all cards
share
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

  Normal Size     Small Size show me how

psych test 33

pierce

QuestionAnswer
null hypothesis is the option that the researcher got the result that they did just by chance. In other words, nothing caused the results to be different from what you’d expect to get just by chance. It was just an accident.
alternative hypothesis that something other than chance caused the results to look they way they do. It’s not just an accident that the results look the way they do. Something made it happen.
alpha level The researcher has to decide just how unlikely the null hypothesis has to be to get them to not believe that it’s true. The odds the researcher decides to use in this situation are referred to as the this for the decision.
expected frequency the frequency you would expect to get by chance
observed frequency the number of times that outcome actually occurred in your data
construct the quality or the aspect of a person that you would like to be able to measure.
true score the place where the person really is on the scale going from the lowest possible score to the highest possible score. The score that we obtain is just a best guess as to what the person’s true score really is.
construct validity The test is measuring the construct it is supposed to measure.
convergent validity The measure is strongly correlated with other measures of the same construct. Example: a new measure of depression is strongly correlated with other established measures of depression.
discriminant validity The new test is uncorrelated with other tests that are clearly designed to measure different constructs.
positive relationship When higher scores on one variable are associated with higher scores on a second variable, we’d say that there is a positive relationship between the two variable
negative relationship it’s a situation where higher scores on one variable tend to go along with having lower scores on the other variable.
no relationship the two variables have nothing to do with each other, as in this case, one’s answer about the direction of the relationship
perfect relationship when the points in the scatterplot fall on a perfectly straight line that’s going either up or down.The variables would tell you everything you needed to know in order to make a perfect guess about their score for the other variable
strong relationship If you run a straight line through the points in the scatterplot and line runs pretty close to those points, but not right through all of them you’d say that you were dealing with this
cross-product When you multiply a person’s deviation score for variable X by their deviation score for variable Y you end up with a value that statisticians refer to as this
covariance the name for the mean of a bunch of cross-products
third variable problem is that the researcher may not have measured scores for this third variable, in which case the researcher has no way of knowing the true cause of the relationship between X and Y.
reliability Is the test consistent? Can we count on the scores from the test?
validity does the test measure what it is suppose to measure?
domain of sampling the full set of information to assess a construct
where does the error come from test construction, test administration, and test scoring and interpretation
test construction the measure may contain info from a different construct and not obtain all the info relevant to the construct
test administrationg differences in test materials, testing conditions, and differences about the examiner
internal consistancy all the items are measuring the same construct
cronbachs alpha average correlation among the items
time-order relationship cause====>effect, cant know which variable occurred first
loaded question includes non-neutral or emotionally laden terms. EX: what is your opinion about the hideous new statue?
leading question attempts to influence the response. EX: shouldnt first year students be aloud to bring a car to campus?
double barreled question ask for more than one piece of information in a single item
open-ended question leaves it up to the subject of info theyre going to give you, not restricting range of responses theyre going to give you. EX: counselor---> ask patient
close-ended ask someone specifically what meds the person takes, give options like 8 meds and they check ones that apply to them, researcher determines possbile responses
partially open-ended "other" is added as an alternate response. questionnaire that says race and gives you a list and at the bottom it says other in case none apply to you.
rating scale not verbal response, rate responses w/ a number.
likert rating scale to what degree does the respondent agrees with the statement. Strongly agree to strongly disagree.
random selection every member of the population has an equal chance of being in the sample
stratified random sample identify various members in a population, each in a different category, make sure the same amount are in each category. 40% males 60% females, you would sample randomly from all males and make sure you have 40 in that sample.
test-retest reliability if you take a test a second time--the scores should vary within the same range as the first time.
split-half reliability compare odd and even scores
Created by: amac