Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
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
Normal Size Small Size show me how
EDL541 Unit 5
| Question | Answer |
|---|---|
| dependability, stability, consistency, predictability, accuracy | reliability |
| 4 types of validity | instrument, internal, external, consequential |
| type of validity: does the test measure what it is supposed to? | instrument validity |
| type of validity: The degree to which observed differences on the dependent variable are directly related to the independent variable, and not to some other (uncontrolled) variable | internal |
| type of validity: The degree to which results are generalizable, or applicable, to groups and environments outside the research settings | external |
| The degree to which correct inferences can be made | validity |
| type of validity: What are the intended and potential unintended consequences of this research | consequential |
| correlations let us know how well we can ___ one score given the other score | predict |
| knowing one piece of information (e.g., the temperature) helps us to better guess what the other piece of data might be (e.g., the beer sales. | associated |
| correlation because both the variables are continuous, or ________, variables | quantitative |
| related or associated because one or both the variables are ________ variables | qualitative or categorical |
| does positively correlated mean one caused the other? | no, just that both go up |
| "r" | Pearson's correlation coefficient |
| high or low correlation? one with an "r" approaching plus or minus 1.0 | high |
| high or low correlation? one with an "r" approaching zero | low |
| the _____ tells us something about how closely two quantitative variables are related | correlation coefficient |
| helps us to increase the accuracy of our predictions of an individual's score on one variable (Y) based on our knowledge of that individual's score on another variable | correlation coefficient |
| T/F The correlation coefficient, even if strong, does not necessarily tell us that large increases in one variable | true |
| T/F The correlation coefficient, even if strong, does not necessarily tell us that large increases in one variable | true |
| T/F The correlation coefficient helps us to explain some of the spread of a group of scores | true |
| Fraenkel and Wallen's suggestion that correlations of about ____ are large enough that they “may have theoretical or practical value | .50 |
| high reliability correlation coefficient from __ to __ | 0.70 - 0.90 |
| appropriateness*, meaningfulness, correctness and usefulness of the inferences a researcher makes | validity |
| 3 types of evidence that an instrument is valid | content related, criterion related, construct related |
| evidence type: format, definition,sample | content related evidence of validity |
| evidence type: relationship between scores | criterion related evidence of validity |
| evidence type: psychological characteristic being measured, hypothetical non observable trait | construct related evidence of validity |
| ____ refers to the consistency | reliability |
| same people getting different scores on same test | errors or measurement |
| 3 ways to test for instrument reliability | test-retest method, equivalent-forms method, internal-consistency method |
| synonym for validity | truthfulness |
| synonym for reliability | consistency |
| procedure for content-related evidence method | obtain expert judgement |
| procedure for criterion-related evidence method | relate to another measure of the same variable |
| procedure for construct-related evidence method | assess evidence on predictions made from theory |
| procedure for test-retest method | give identical instrument twice |
| procedure for equivalent-forms method | give two forms of instrument |
| procedure for equivalent-forms/retest method | give two forms with time interval between |
| procedure for internal consistency method | divide instrument into halves and score each |
| procedure for scoring observer agreement method | compare scores obtained by two or more observers or scorers |
| it is important to use ___ instruments for the conclusions are based on the info obtained from the instruments | valid |
| a ___ is a numerical index representing the degree of correspondence between scores on an instrument and criterion measure | validity coefficient |
| the degree that observed differences on dependent variable are directly related to the independent variable no to some extraneous variable | internal validity |
| individuals or groups differing form one another in unintended ways (age, strength, political beliefs) | subject characteristic threat or selection bias |
| losing some of the subjects as the study progresses | mortality threat |
| the classroom or environment that the study takes place in | location threat |
| the way in which an instrument is used | instrument threat |
| scorer may unconsciously distort the data | data collector bias |
| taking a pretest may influence post test results | test threat or test-retest familiarity |
| unplanned events that occur during the study | history threat |
| changes over course of time rather than independent variable | maturation threat |
| how subjects view the study | attitude effect |
| positive effect from increased attention to subjects, type of attitude effect | Hawthorne effect |
| a group selected for its unusually high or low pre intervention will naturally score closer to the mean | regression threat |
| who or how the treatment is administered | implementation threat |
| major purpose of correlational research | clarify understanding by identifying relationships among variables |
| another purpose of correlational research if the relationship is of significant magnitude | predict a score on one variable if the score on the other is known |
| straight line made on a scatterplot | regression line |
| two or more predictor variables in a correlational study then we use: | multiple regression |
| basic steps in correlational research | problem selection, sample, instrument, design and procedures, data collection, data analysis and interpretation |
| this is somewhere between 0 and +1 or 0 and -1 | correlational coefficient |
| the closer the correlational coefficient is to __ the stronger the relationship | 1 (plus or minus) |
| a problem with Pearson's r, the common correlational coefficient | only useful for linear relationships |
| T/F make sure to look at scatterplots in correlational studies not just correlational coeffiecients | true |
| *The ability to measure what it says it measures refers to: | instrument validity |
| *To question whether the cause was what the researcher claimed was the cause is to question the study's: | internal validity |
| *When researchers ask what intended and unintended things might happen as a result of their research or evaluation efforts, they recognize the issue of: | consequential validity |
| * To ask, "Can we generalize from our sample to the population we claim this sample represents?" is to begin to address the issue of: | external validity |
| *If we know there is a positive correlation between temperature and ice cream sales, we know that, in general: | both go up |
| * If there is a negative correlation between number of hours worked and grade point average (GPA), we would know that, in general | one up one down |
| * If we had a correlation of r=.90, we would have a: | strong positive correlation |
| *t/f If all we know is the group's mean, it is often the best guess we have of any individual's score. That is, even though our guess won't be exactly right most of the time, we're generally better off guessing the mean than any other number. If we fi | true |
| *t/f A correlation of r=.90 indicates that, when the first variable goes up a modest amount, the second variable goes up a great deal. | false |
| *t/f When we find a correlation between two variables, such as time spent reading and test scores, we can say with confidence that the time spent reading caused the higher test scores. | false |
| *_______ refers to the consistency of the scores obtained. | reliability |
| *Allowing a time interval to elapse between administration of an instrument (such as a science aptitude test) and obtaining the criterion scores (such as the next year's grade in science) is a way to obtain evidence of: | predictive validity |
| * When instrument data and criterion data are gathered at nearly the same time, and the results compared, this is an attempt by researchers to obtain evidence of | concurrent validity |
| *hen two different but equivalent forms of an instrument are administered to the same group of individuals during the same time period, this form of reliability is called: | the equivalent-forms method |
| *When the same test is adminstered twice to the same group of people after a certain time interval has elapsed, this method of reliability is called the: | test-retest method |
| * When subjects in a study are likely to be lost due to any number of reasons, such a moving away from a district, it is referred to as: | mortality threat |
| *t/fAccording to Fraenkel and Wallen, one way to obtain content-related evidence of validity is to have someone who can render an intelligent judgement about the adequacy of the instrument look at the content and format of the instrument and judge whether | true |
| *t/f If educational researchers gave the same people the same test on two dates a few days apart, they would expect the scores to come out exactly the same | false |
| *The positive effect resulting from increased attention and recognition of subjects just from being part of a study has been referred to as the"Hawthorne effect." (Informally, this term is used when a group does better just because they knew they were in | true |
| *In correlational research, researchers investigate possible relationships among variables without necessarily trying to influence those variables | true |
Created by:
Mrs. Bridgman