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 4
| Question | Answer |
|---|---|
| 4 things you should take notes on from research | who wrote it, the problem, the procedures, the findings and conclusions |
| When reading research you are looking at ___ the researcher studied, ___ they studied it, and _____ | what, how, what they found out |
| The challenge in reading research from a critical angle is so you will neither ___ or ____ it | trust it blindly or dismiss it for being imperfect |
| 2 questions to ask yourself: | did the study investigate what they said they would, did the conclusion match the findings |
| in the example of hospitals and schools the ___ data was misinterpreted | descriptive |
| Are the __, __ and __ of the study talking about the same thing? | beginning, middle, end |
| how to read in a hurry questions: | what was the conclusion, are the data consistent with that conclusion |
| ___ studies canhelp us better understand attitudes of stakeholders, what actually happens in schools, but don't tell us about relationships among variables | descriptive |
| ___ studies are helpful because it shows us relationships among variables but cannot tell us if the relationship is causal | associational |
| ___ studies (especially ____ studies)can give us the highest confidence that one variable caused an effect in another variable | intervention-type, experimental |
| According to F&W, formal experiments consist of these 2 conditions: | two methods are compared to assess effect of independent variable, the independent variable is manipulated |
| the essential ingredient in true experimental design is that subjects are ____ to treatment | randomly assigned |
| first thing to ask yourself about causal relationships | were subjects randomly assigned |
| if subjects not randomly assigned should we dismiss results? | no, descriptive info still stands |
| if we cannot randomly assign we use: | quasi experimental or associational research |
| the benefit of random assignment is that all of the known and unknown ____ are randomly and equally distributed | extraneous variables |
| protocol for assessing causal relationships: | is the claim based on true experimental research? and did the researcher randomly assign INDIVIDUAL subjects? if not was it quasi experimental or associational? |
| are control and comparison groups synonymous? | no, control receives no treatment or placebo, comparison receives different kind of treatment |
| what is z score of student who scores 10 if the mean is 10? | zero |
| z score is equal to the difference between individual scores divided by: | the standard deviation |
| if standard deviation is 2, mean is 10 and raw score is 14, what is student's z score | +2 = (14-10)/2 |
| z score of ___ is within 1 standard deviation of the mean | 1 |
| ___% of scores are within 1 standard deviation of the mean | 68 |
| if Joe has a z score of zero, what does that tell you? | Joe's score was average (zero points from mean), about 50% scored higher |
| Mary had a z score of +1 what does that tell you? | Her score is above the mean, one standard deviation above the mean, she did better than 84% of others |
| Pat had a z score of -1 what does that tell you? | His score is below the mean, one SD below, he did better than 16% of others |
| the most common type of T Test | test for differences between means (average score of treatment and control groups) |
| if t test finding is that the difference is not significant it means: | differences not large enough to be confident to generalize findings to larger population |
| if t test finding is statistically significant it means: | there is a difference, but we don't know how large |
| can two studies with same t test difference disagree on statistical significance? | yes, depends on overall population size |
| t tests workhorse piece is | difference between means |
| t test difference between means is adjusted based on: | sample size |
| the bigger the difference between means, the bigger the sample size and the less spread out the scores the ___ likely it is to be statistically significant | more |
| ____ refers to the process of selecting the individuals from whom the researcher will obtain data for his study | sampling |
| a __ is a group from which info is obtained and the ___ is the group to which it is applied | sample, population |
| before selecting a sample we must determine whom we would like to __ results of study | apply |
| ____ can be any size and will have at least one characteristic that sets it apart | population |
| populations are people T/F | false could be classrooms |
| the actual population you'd like to generalize | target population |
| the population you are able to generalize | accessible population |
| population and sample should be described in detail so that: | interested parties can determine if info is applicable |
| two types of sampling: | random vs nonrandom |
| in a simple random sample each member of the population has an ___ chance of being selected | equal |
| three most common random sampling methods | simple random, stratified random, and cluster sampling |
| each and everyone in population has equal chance in this type of random sample | simple random sample |
| this type of random sample certain subgroups or strata are selected in the same proportion as they exist in the general pop | stratified |
| this type of random sample involves the selection of groups rather than individuals | cluster |
| three types of non random sampling | systematic, convenience, purposive |
| type of non random sample: every nth individual is selected | systematic |
| type of non random sample: group of individuals who are available for study | convenience |
| type of non random sample: specifically select individuals who will provide the info they need | purposive |
| a problem with systematic sampling if population is already ordered in a way that coincides with selection | periodicity |
| how is table of random number used? | assign each member of population a number, use random table to choose numbers |
| sample sizes for descriptive and corelational | 100, 50 |
| sample sizes for experimental and causal-comparitive | 30 per group |
| sample sizes for qualitative | 1-20 |
| the extent that the results can be generalized from a sample to a population | external validity |
| using random samples and inferential statistics to generalize | population generalizability |
| the findings of one study can apply to other settings | ecological generalizability |
| two ways to improve validity when using non random samples | detailed description of sample and replication |
| the device used to gather data | instrument |
| why need to know the who, what, when, why... of data collection? | affects the data that is collected |
| a valid instrument is one that: | measures what it is supposed to measure |
| a reliable instrument is one that: | gives us consistent results |
| who can provide data? | researchers themselves, the subjects of the study, informants |
| others who are knowledgeable about the subjects of the study | informants |
| tally sheets, field notes, or recorded observations are types of ___ instruments | researcher |
| tests or questionnaires completed by subjects are ____ instruments | subject |
| interview or questionnaires completed by people who know the subjects are ___ instruments | informant |
| two ways to acquire an instrument: | use a standard , create one |
| tests, questionnaires, rating scales are examples of ____ instruments whereas ____ instruments measure a procedure or product | written response, performance |
| the ___ of creating and instrument needs to be understood before interpreting the ____ | process, product |
| to eliminate the reactive (negative) effect on subjects, researchers use ____, procedures that involve no intrusion into normal activities | unobtrusive measures |
| the group to which the comparison is made | norm group |
| instruments that provide info on a comparison group | norm referenced instrument |
| instruments that focus on instruction, goal, or target for a learner to achieve | criterion referenced instrument |
| percentile ranks ____ be used by classroom teachers to compare students' abilities and ___ be used by researchers to calculate means of a group | should, should not |
| simplest form of measurement, assign numbers to categories | nominal scale |
| form of measurement: data is ordered in some way | ordinal scale |
| form of measurement: data is ordered and the distances between points are equal | interval scale |
| form of measurement: data is ordered, equal distances, and contains a zero point | ratio scales |
| ___ & ___ scales provide more info than the other two | ratio and interval |
| some type of ___ procedures are inappropriate for different scales | statistical |
| type of scale used to show different amounts of money | ratio (ordered, increments, zero) |
| type of scale used to show temperature range | interval (ordered, increments) |
| type of scale used to show position in a race | ordinal (ordered) |
| type of scale used to show two categories like male and female | nominal (assigned a number) |
| the only type of research that directly attempts to influence a particular variable | experimental |
| the best type of research to test cause and effect relationships | experimental |
| experimental research often looks at the ___ of at least one ___ variable on one or more ___ variables | effect, independent, dependent |
| the major characteristic of experimental research is that the researcher ___ the independent variable | manipulates |
| examples of independent variables in ed research | method of instruction, type of assignments |
| examples of dependent variables in ed research | interest, attention span |
| manipulation of independent variable consists of two levels: | the treatment is present and the treatment is not present |
| variables that cannot be manipulated | gender, ethnicity, religion |
| equal chance of being assigned to experimental or control group | random assignment |
| random assignment begins: | before experiment begins |
| random assignment allows equivalent groups so that they only differ by: | chance |
| number of subjects in experimental groups | 40 each (no less than 20) |
| ways to control extraneous variables | randomization, holding variables constant, matching |
| randomized posttest only control group is known as the: | prototypical "true experimental" design |
| type of design: involves repeated measurement before and after treatment | time series design |
| type of design: involves two independent factors | factorial design |
| whether a cause is due to what the researchers said it was or something else | internal validity |
| threats to internal validity | mortality, location, instrumentation, testing, history, attitude... |
| *If one wished to investigate whether one thing (such as a specific teaching method) caused another thing (such as higher test scores), the ideal would be to create a study that used an _________ design because it is the best design for investigating caus | experimental |
| *Calculations for the effect size and for the t-test used to compare means have at least one thing in common: the workhorse piece of the calculations for both is the difference between two __________. (On the other hand, the rest of the calculation for ea | means |
| *Ms Smith, a high school teacher, has students whose standardized test scores are much higher, on average, than those of Mr Jones's high school students. While we can't tell for sure from just this piece of information, which of the following should be ou | The students Ms Smith and Mr Jones work with differ. |
| *According to the professor, when reading a piece of research, it is important to note (and take notes on) which of the following? | all of the choices |
| *A study reports testing a direct instruction method for teaching language arts students to recognize the difference between direct objects and indirect objects. The study notes that 2 classrooms were randomly assigned to either the direct instruction or | quasi-experimental |
| *A study reports that male students' GPA's in a Block 8 school are lower than female students' GPA's in that school. Which research design is this? | associational |
| *hich statistic is better for indicating whether a difference between the means of two groups (such as between the cooperative learning group and the traditional group) is a large difference? | effect size |
| *What one evaluator referred to as the "Sesame Street Problem" is the situation that: | Given a large enough sample size, very small effects can be declared statistically signifiant |
| *A finding of a significant difference between two groups (as between the technology-enhanced lessons and the traditional lessons) tells us which of the following? | That, in generalizing our results to the population from which our samples were drawn, we can be relatively confident that the difference between the groups is greater than zero |
| *A "significant difference" is always a large difference. | false |
| *If one study reports finding a significant difference between two methods of instruction, and another study on the same topic reports that there was not a significant difference between those two methods of instruction in their study, we can, with a reas | false |
| *Technically, instead of referring to "population statistics," we should refer to the numbers that describe our sample as "statistics" and the numbers that describe the population as "parameters." | true |
| *If a study finds that there was no significant difference between two methods of teaching, we can be certain that the two methods are equally effective. | false |
| *In a stratified random sample, the researcher selects individuals from "strata," or _____________________, instead of randomly from the population as a whole | subgroups |
| *The degree to which results are generalizable, or applicable, to groups and environments outside the research setting is called ________________ validity. | external |
| *__________________ (lower-case) instruments can measure either a procedure or a product. | performance |
| *The group from which information is obtained is called the: | sample |
| *The group to which the results of the study are intended to apply is the: | population |
| *The population from which the researcher can realistically select subjects for a sample, and to which the researcher is entitled to generalize findings is called the: | accessible population |
| *If every member of the population has an equal chance of being selected, then the sample is: | random |
| *When every "nth" individual in the population is selected for inclusion in a sample, this is called: | systematic sampling |
| *When the sample consists of people who were in a group that was available for study (such as a classroom or set of classrooms), this is called: | convenience sampling |
| *When researchers use their judgement to select a sample that they believe will provide the data they need based on prior information, such as when researchers choose specific members of a community to participate in a focus group because those community | purposive sampling |
| *What are three types of instruments that can be used to obtain data? (These are based on who collects the data.) | informant instruments, researcher instruments, and subject instruments |
| *In Dr. Matthews' opinion, creating a new instrument is the best choice when: | the goal of the study is to discover something new or to test for something that's not measured on standardized tests. |
| *A scale which uses numbers to rank data is a(n): | ordinal scale |
| *Populations must be large. | false |
| *In order to define a population, it is necessary for the researcher to ask, "To whom do I want the results of the survey to apply?" | true |
| *In Dr. Matthews' opinion, obtaining the largest sample size possible is always the best way to get an accurate picture of the results. | no, It is not always best to have a large sample size if the return rate of the sample is not high. With a smaller sample size, researchers can do more follow up. |
| *The term "data" refers only to information which is quantifiable. | false |
| *A valid instrument - according to the old-fashioned but still good definition - is one that measures what it says it measures. | true |
| *A Likert scale can provide easily summarizable information. | true |
| *Average-and-below students taking standardized tests will, by design, often have to miss one-third to one-half or even more of the questions on the standardized test. | true |
Created by:
Mrs. Bridgman