Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
180 Cards in this Set
- Front
- Back
Measure of central tendency for Nominal level of measurement
|
mode
|
|
Measure of central tendency for ordinal level of measurement
|
mode & median
|
|
Measure of central tendency for interval level of measurement
|
mode, median & mean
|
|
Measure of dispersion for Nominal level of measurement
|
range
|
|
Measure of dispersion for ordinal level of measurement
|
range, interquartile & semiquartile range
|
|
Measure of dispersion for interval level of measurement
|
range, interquartile & semiquartile range, standard deviation
|
|
What is a measure of central tendency?
|
a mathematical property of a distribution, which describes what is typical of that distribution
|
|
What are descriptive statistics?
|
information about the typicality of a distribution as well as the dispersion of values and the shape of the distribution
|
|
modal category
|
the most frequently occurring category within a distribution of values (e.g., males
in the class) |
|
mode
|
is the frequency associated with that category (e.g., 23 males)
|
|
median
|
the scale value that divides the top 50% of a distribution from the bottom 50% of a distribution
|
|
3 methods of finding the median
|
1. counting until you find the middle
2. create a frequency distribution & find the 50th percentile 3. calculate: use this when there are tied ranks & when scale is continuous |
|
equation to calculate median
|
median = lower limit + (range * (.5 N - # below lower limit/# cases in the interval))
|
|
Tied ranks
|
two or more of the same value occurring in a distribution.
|
|
exact lower limit
|
in a frequency distribution, the value at the midpoint between the
value of that category and the one below it |
|
arithmetic mean
|
the arithmetic average of all the scores in a
distribution |
|
deviation score
|
the distance, either positive or negative, that a value lies above or below the mean of
the distribution. A deviation score is often symbolized as d |
|
advantages of using the mode as a measure of central tendency
|
1. easy to compute
2. most flexible & can use for nominal 3. least effect of extreme scores |
|
advantages of using the median as a measure of central tendency
|
1. can be used for O, I, R
2. less sensitive than mode to groupings into class intervals 3. less affected by extreme scores than mean |
|
advantages of using the mean as a measure of central tendency
|
1. reflects actual value of scores
2. most stable measure of CT 3. can be manipulated algebraically |
|
disadvantages of using the mode as a measure of central tendency
|
1. there may be more than one
2. sensitive to size and number of class intervals e.g. $0-24K, $25-49 -> changing the size of the interval will move the mode across categories 3. unstable |
|
disadvantages of using the median as a measure of central tendency
|
1.hard to manipulate
2. less used for inferential statistics |
|
disadvantages of using the mean as a measure of central tendency
|
1. affected by extreme scores
2. can only be used for I or R scales |
|
What are the relative positions of the mean, median and mode in a positively skewed distribution?
|
mode, median & mean -> long tail to the right
|
|
Sum of the deviation scores
|
The mean of a distribution is the point at which the sum of the deviation scores from the mean equals 0
|
|
range
|
-the difference in frequency between the most frequent and the least frequent category
-a measure of dispersion -particularly suited for nominal-level data |
|
interquartile range
|
-the difference in scale value between the 75th percentile and the 25th percentile
-particularly useful for ordinal data |
|
semi-interquartile range
|
-1/2 of the interquartile range
-describes the 25% of values closest to the median -aka quartile deviation |
|
When should you use the interquartile or semi-quartile range?
|
-when distribution is skewed by extreme scores
-when data are ordinal -don't use for inferential stats -use with median to describe the dispersion |
|
Name appropriate measures of dispersion for the NOIR levels.
|
N - range
O - interquartile/semi-quartile range I - mean square (variance), SD R - variance, SD |
|
sum of squares
|
the sum of the squared deviations of each value in a distribution from its mean
|
|
mean square
|
-the average, or mean, of the sum of squares
-aka variance |
|
standard deviation
|
-square root of the mean square (variance)
-converts squared units into back into scale units |
|
when to divide by N-1
|
-when estimating the population variance
|
|
Degrees of freedom
|
-number of independent pieces of data that are available for inference to a population
-in samples, it is symbolized as N – 1. |
|
biased estimator
|
- variance and standard deviation of a sample
-does not generalize well to a population (i.e., does not estimate well), but correctly describes a sample |
|
why is it important that the mean of a sample is a good estimator of the population mean?
|
-you don't have to adjust it when estimating for the population, that is, you NEVER divide by N-1
|
|
skewness
|
used to describe distributions that “lean” towards the upper end of a scale (i.e., negative skew) or towards the lower end of a scale (i.e., positive skew)
|
|
kurtosis
|
describes “flatness” or “peakedness” in a distribution
|
|
platykurtic
|
a very flat distribution
|
|
leptokurtic
|
a very peaked distribution
|
|
mesokurtic
|
between platy- and leptokurtic
|
|
What are the 4 "moments around the mean?"
|
-four statistical measures that describe the mathematical characteristics of a distribution
1. average deviation - defines position of the mean 2. variance - spread of values around the mean 3. skewness - sum of cubed deviations 4. kurtosis - (avg dev_^4) |
|
What does negative kurtosis mean?
|
tall, peaked (leptokurtic) distribution > the more positive, the flatter it is
distributions with negative kurtosis tend to show less variability |
|
What are the three methods of determining the IQR?
|
1. IQR = Q3-Q1
2. once you know the median, count to the medians of the distribution below and above 3. calculate exact midpoints above & below median using formula |
|
steps of research (3 pts)
|
1. Pose question;
2. Collect data; 3. Present answer to the question |
|
importance of educational research (3 pts)
|
1. adding to knowledge (original or replication);
2. improving practice (suggesting improvement, evaluating approaches, building connections with other practitioners/researchers); 3. informing policy |
|
pitfalls in research
|
1. vague/contradictory findings;
2. questionable data (inappropriate sample, instruments, analysis); 3. unclear intent; 4. lack of full disclosure |
|
6 steps in research process
|
1. identify problem (specify, justify);
2. review literature (select & summarize sources); 3. specify purpose (identify and narrow purpose statement to research questions); 4. collect data (select sample, obtain permission, get info) ; 5. analyze & interpret data (represent & explain data); 6. report & evaluate research (choose audience, structure & write report) |
|
quantitative research: 6 steps
|
1.research problem: description of trends, need for explanation of relationship between variables;
2.lit review: suggest research questions & justify need; 3.purpose: specific, narrow, measurable & observable; 4.data collection: numeric, from large N using instruments with present questions & answers; 5.analysis: trnds, comparing groups, relating variables statistically,comparing results to prior predictions and past research; 6.report: standard, fixed structure & evaluation criteria, objective |
|
qualitative research: 6 steps
|
1.research problem: exploratory, developing detailed understanding of a central phenomenon;
2.lit review: minor role, justifies problem; 3.purpose: general, broad to be inclusive of participants' experiences; 4.data collection: based on words from small number of individuals, protocols developed to record data as study proceeds (e.g., observational, interview, transcribed audio); 5.analysis: data analyzed for description and themes using text analysis and interpreting larger meaning of the findings; 6.report:flexible, emerging structures and evaluative criteria, includes researchers' subjective reflexivity & bias |
|
quantitative research designs (3)
|
experimental (intervention/group comparison studies),
correlational, survey design |
|
qualitative research designs (3)
|
grounded theory designs (based on model developed from interview data),
ethnographic design (one group of individuals in setting where they live & work, examines how they interact), narrative research (describes lives of individuals) |
|
mixed methods designs
|
both quant & qual. to provide better understanding of the problem, may be multiphase or combined into a single study;
need to decide on priority given to each type of data, how to integrate data and whether to use theory to guide study |
|
action research designs
|
like mixed but focus on practical problems in schools & classrooms
systematic procedures used by teachers/others to address improvements in setting, teaching & students' learning |
|
ethical issues in conducting research
|
institutional review boards - study must conform to guidelines
professional associations - often address ways to give back to participants ethical data collection - minimize disruption to site (too much time, interference with ongoing work, too late in year) - 'gatekeepers' help, risk to participants reporting data - report honestly, don't plagiarize, give credit where due |
|
research skills
|
solving puzzles, attention span, library skills, writing & editing
|
|
research problem versus topic
|
topic is broad subject matter, the problem is narrower issue
|
|
research problem versus purpose of study
|
purpose is major objective of study that addresses the problem
|
|
research problem versus research questions
|
questions narrow the purpose into specific questions to be answered in study
|
|
5 questions to assess whether to research a problem
|
1. Is there a gap in the literature?
2. Does it replicate a past study with a new sample & site? 3. Does the study extend past research? 4. Does the study give voice to marginalized individuals? 5. Does the study inform practice? |
|
(4 pts) When quantitative may be more appropriate
|
1. measure variables
2.assessing impact of variables on outcome 3.testing theories or broad explanations 4. applying results to large number of people |
|
(4 pts) When qualitative may be more appropriate
|
1. learning about individuals' views
2. assessing process over time 3. generating theories based on participant perspectives 4. obtaining detailed information about a few people or research sites |
|
(5 pts) Parts of a 'Statement of the Problem' section
|
1. topic (broad subject matter)
2. research problem 3. justification of importance based on past research 4. gaps in existing knowledge 5. audiences that will benefit from study |
|
narrative hook (first sentence of Statement of Problem)
|
May contain:
1. statistical data 2. provocative question 3. clear need for research 4 intent or purpose of study -begin with introduction that is easy to understand and open with statement that draws reader interest |
|
strategies for writing "Statement of the Problem" sections
|
1. use a template
2. cite literature often 3. reference statistical information and quoates |
|
purpose statement
|
sentence in study that states overall direction/objective
|
|
research question
|
questions that focus purpose of study into specific areas
|
|
hypothesis
|
statement that narrows purpose statement into specific predictions about relationships among variables
|
|
research objective
|
statement of intent used in quantitative research that specifies goals, often subdivided into major and minor objectives, often appear at end of 'statement of the problem' sections, after lit review
|
|
variable
|
measurable and variable characteristic or attribute of an individual or organization
|
|
measurement
|
researchers records information from individuals - self-report questionnaire, observation and recording of scores
|
|
categorical vs. continuous variables
|
categorical = discrete/nominal (measured as a small number of groups/categories)
continuous = interval/rating/scaled score (measured as a point on a continuum) |
|
construct
|
attribute or characteristic stated in a specific, applied way (e.g. achievement is construct, GPA is a variable)
|
|
4 questions to distinguish between types of variables
|
What am I trying to explain? (dependent)
What variables influence the outcomes? (independent) What else do I need to measure to make sure the variables in question influence outcomes? (control and mediating) What variables influence outcomes bu cannot/will not be measured? (confounding variables) |
|
dependent variable
|
attribute influenced by the independent variable
|
|
independent variable
|
attribute that influences/affects outcome or dependent variable
|
|
types of independent variables (4)
|
measured, control, treatment, moderating
|
|
measured variables
|
independent variable that is measured in a study, used in experiments and survets e.g. age of a child
|
|
control variable
|
(covariate) IV that is of secondary interest and neutralized through statistical or design procedures, used in experiments, correlational studies e.g. demographic info like age, gender, SES
|
|
treatment variable
|
IV that is manipulated by the researchers, categorical & composed of 2+ groups, used in experiments
|
|
moderating variable
|
IV that is of secondary interest that combines with another IV to influence the DV, measured & observed as it interacts with other variables, used in experiments, e.g. demographics, measured variables like attitude, manipulated variable like classroom instruction
|
|
where to locate an independent variable
|
in purpose statements, research questions & hypotheses: look for variable that influences or predicts an outcome
|
|
intervening variable
|
attribute or characteristic that "stands between" IV and DV and influences DV independently - mediates the effecrs of the IV
|
|
confounding variable
|
aka spurious variables = attributes the research can't directly measure even though the influence the relationship between IV and DV
|
|
probable causation
|
researchers attempt to establish cause-and-effect relationships rather than prove the relationship
|
|
a theory in quantatitive research
|
explains and predicts the probable relationships between IV and DV
|
|
quantitative purpose statement
|
identifies variables, relationship, participants and site for research
|
|
basic steps in writing quantitative research questions
|
-pose question
-begin with how what or why -specify IV DV mediating& control variables -describe/compare/relate to indicate connection between variables -indicate participants and research site |
|
descriptive question
|
identifies participants' responses to a single variable/questions
|
|
relationship question
|
tries to identify degree and magnitude of relationship between two or more variables
|
|
comparison question
|
ask to find out how 2+ groups on an IV differ in terms or outcome variables
|
|
null hypotheses
|
predicts that there is no relationship between independent and dependent variables. no difference between groups of an IV or DV
|
|
alternative hypothesis
|
not the null hypothesis, suggests other possibilities based on results from past research or explanation from the literature
|
|
directional alternative hypothesis
|
predicts the direction of a change in the total population of people
|
|
non-directional alternative hypothesis
|
predicts a change, a difference or a relationship for variables in a population but does not indicate positive or negative direction
|
|
central phenomenon
|
concept/process explored e.g. ethnic identity of Chinese American immigrants
|
|
differentiating between quant and qual
|
quant,qual: hypotheses vs. research questions , identifies and tries to measure multiple variables vs. central phenomenon, testing theories that predicts results from relating variables vs. non testing theories, close-ended vs. open-ended where phenomenon of interest may change, measures differences and magnitudes of differences or chnge over time vs. looking for views of small group or individuals
|
|
emerging process
|
intent/purpose & questions researched may change based onf feedback/responses from participants
|
|
qualitative purpose statements
|
indicates intent to explore/understand central phenomenon with specific individuals at certain site, typically at end of introduction
|
|
strategies for writing central questions
|
begins with how or what, specifies central phenomenon, identifies participants, mentions research site
|
|
sub-questions
|
refines the central questions (more specific)
|
|
issue subquestions
|
narrow the focus of the central question into specific questions
|
|
procedural subquestions
|
indicate steps to be used in analyzing data (e.g. which question will be answered first, orom identifying categories to tracing a specified process)
|
|
distinguishing qual from data collection questions
|
in data collection, you also ask participants about themselves and ask them to suggest other helpful individuals
|
|
research problem versus topic
|
topic is broad subject matter, the problem is narrower issue
|
|
research problem versus purpose of study
|
purpose is major objective of study that addresses the problem
|
|
research problem versus research questions
|
questions narrow the purpose into specific questions to be answered in study
|
|
5 questions to assess whether to research a problem
|
1. Is there a gap in the literature?
2. Does it replicate a past study with a new sample & site? 3. Does the study extend past research? 4. Does the study give voice to marginalized individuals? 5. Does the study inform practice? |
|
(4 pts) When quantitative may be more appropriate
|
1. measure variables
2.assessing impact of variables on outcome 3.testing theories or broad explanations 4. applying results to large number of people |
|
(4 pts) When qualitative may be more appropriate
|
1. learning about individuals' views
2. assessing process over time 3. generating theories based on participant perspectives 4. obtaining detailed information about a few people or research sites |
|
(5 pts) Parts of a 'Statement of the Problem' section
|
1. topic (broad subject matter)
2. research problem 3. justification of importance based on past research 4. gaps in existing knowledge 5. audiences that will benefit from study |
|
narrative hook (first sentence of Statement of Problem)
|
May contain:
1. statistical data 2. provocative question 3. clear need for research 4 intent or purpose of study -begin with introduction that is easy to understand and open with statement that draws reader interest |
|
strategies for writing "Statement of the Problem" sections
|
1. use a template
2. cite literature often 3. reference statistical information and quoates |
|
Martin Seligman's Measure of Explanatory Style predicted retention rates of college freshmen better than students' SAT scores. Seligman's instrument is said to have ____ validity.
|
Criterion-related validity
|
|
Obtaining the same results on repeated administration of the same instrument is known as:
|
test-retest reliability
|
|
A researcher who is interested in the effects of low birth weight (LBW) on subsequent kindergarten readiness follows a group of LBW children through the first five years of life, collecting various psychological, social, and educational outcomes. She correlates all of these variables with the original birth weight of the child. The measure of birth weight is:
|
ratio
|
|
A researcher is interested in a construct she calls "school satisfaction." To measure this, she asks schoolchildren in grades 5-12 to fill out a 10-item survey about school experiences (e.g., "I feel challenged at school."). Students complete the questions on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). This measure of school satisfaction is:
|
interval
|
|
A researcher is interested in a construct he calls "school satisfaction." To measure this, he asks schoolchildren in grades 5-12 to circle either "yes" or "no" on a list containing 10 adjectives describing school (e.g., interesting, challenging, boring). This measure of school satisfaction is:
|
nominal
|
|
The number of aggressive acts on a playground observed during a school-day recess period is an example of a(n):
|
operational definition
|
|
With increasing sample size a researcher gains:
|
less sampling error
|
|
Choosing a sample by selecting participants because they are available and willing to participate in the study is known as:
|
convenience sampling
|
|
In a study of interscholastic athletics on academic achievement, a researcher first identifies state high school athletic associations in all 50 states. From this list, she develops the list of all secondary schools offering at least 12 interscholastic sports (6 girls, 6 boys). From that list of schools, she randomly selected 435 high schools across the country proportional to the population. (She does this by picking one high school from each congressional district.) From that list, she randomly selects the participants from one boys sport and one girls sport in each school. The result is a mailed survey to over 13,000 high school students participating in interscholastic athletics. This approach to sampling is known as:
|
multistage cluster sampling.
|
|
Choosing a sample by partitioning off the population based on selected characteristics (e.g., race) such that each characteristic is represented proportionately in the population is known as:
|
stratified sampling
|
|
Using some research participants to help you identify other possible participants in the population is known as:
|
snowball sampling
|
|
Choosing every "nth" individual or site in the population until the desired sample size is achieved is known as:
|
systematic sampling
|
|
Choosing a sample such that every member of the population has an equal chance of being selected is known as:
|
simple random sampling
|
|
Nonprobability sampling is characterized by choosing a sample based on:
|
availability of participants
|
|
The difference between the sample estimate and the true population score is the
|
sampling error
|
|
Compared to a summed scale, what is a disadvantage of a single-item score?
|
may not reliably and accurately reflect an individual's score
|
|
Which is the flow of data analysis as described in the text?
|
inputting, cleaning, assessing missing data, computing descriptive statistics
|
|
One way to handle missing data is to eliminate any participant who has at least one missing data point. The main disadvantage to this approach is that it:
|
reduces the size of the data set
|
|
What would be the percentile score of a student whose standardized (z-score) on the SAT is +2?
|
99
|
|
What would be the percentile score of a student whose standardized (z-score) on the SAT is –1?
|
16
|
|
What statistical test would be used to test mathematics achievement scores between students learning from the following three different pedagogical strategies: induction, theory only, theory + examples?
|
analysis of variance
|
|
Type II error
|
false negative - probability = beta
|
|
Type I error
|
false positive - probability = alpha
|
|
Measures of central tendency
|
mean, median, mode - value or score that represents the entire distribution
|
|
Measures of variability (describes the “spread” of the scores
|
range, SD
|
|
Measures of relative standing
|
percentile rank, calculated score like Z score
|
|
types of inferential statistics (5)
|
t-test, ANOVA, chi-squared, pearson correlation, multiple regression
|
|
Hypothesis testing:
|
A procedure for making decisions about results by comparing an observed value of a sample with a population value to determine if no difference or relationship exists between the values
|
|
Confidence interval:
|
The range of upper and lower statistical values that is consistent with observed data and is likely to contain the actual population mean
|
|
effect size
|
A means for identifying the practical strength of the conclusions about group differences or about the relationship among variables
|
|
things to consider when selecting a statistic
|
scale of measurement for Ivs and DVs, skew/normality of distribution
|
|
power
|
probability of a correct rejection
|
|
how many tables when reporting stats
|
one per test
|
|
major steps of discussion
|
Summarize major results
Explain why they occurred Explain the implications of the results for the audiences Advance limitations Suggest future research End on positive note |
|
grounded theory research
|
systematic, qualitative procedure used to generate a theory to explain a process, action or interactions about a substantive topic at a broad conceptual level
|
|
types of grounded theory designs (3)
|
systematic, emerging design, constructivist design
|
|
key characteristics of grounded theory research (3)
|
process approach
theoretical sampling constant comparative data analysis core category theory generation memos |
|
ethical issues in conducting grounded theory research
|
-not advancing purpose of the study
-creating power & authority imbalances in interviewing -not building useful chain of evidence -not engaging in a study that benefits participants |
|
steps in conducting a grounded theory study
|
1. decide if GT design is best for research problem
2. identify process to study 3.seek approval & access 4. conduct theoretical sampling 5. code data 6. use selective coding & develop theory 7.validate theory 8. write GT research report |
|
ways to evaluate grounded theory study
|
1.check for obvious connection between categories and data
2.chck if theory is useful conceptual explanation for process 3. ask if theory provides relevant explanation of actual problbems and basic process 4. check if theory can be modified as conditions change or more data arrives 5.ask if theoretical model was developed 6. ask if there is a central phenomenon or core category 7. ask if model emerges through phases of conding 8. ask if research tries to interrelate categories |
|
systematic design
|
open coding, axial coding, selective coding
|
|
open coding
|
properties & dimensionalized properties
|
|
axial coding
|
researchselects one open coding category & examines it as the central phenomenon, then relates all other categories to it
|
|
selective coding
|
writing a theory based n the interrelationship of the categories from axial coding
|
|
emerging design
|
grounded theory exists at most abstract conceptial level, theory is not forced into categories, 4 essential critera are fit, work, relevance & modifiability
|
|
constructivist design
|
philosophical position between positivist and postmodern researchers, theorist explains individuals' feelings as they experience a process/phenomenon, mentions beliefs & values of reseach and does not use predetermine categories, narrative is explanatory and probes assumptions and meanings for studied individuals
|
|
process approach
|
sequence of actions and interactions among people and events pertaining to a topic
|
|
Theoretical sampling
|
The researcher chooses forms of data collection that will yield text and images useful in generating a theory
|
|
Constant comparative data analysis:
|
An inductive (from specific to broad) data analysis procedure in grounded theory research of generating and connecting categories by comparing incidents in the data to other incidents, incidents to categories, and categories to other categories
|
|
A core category
|
A category that can become the theme that describes or becomes the main theme of the process
|
|
theory generation
|
creation of an abstract explanation or understanding of a process about a substantive topic grounded in the data
|
|
memos
|
notes written by researcher through research process to elaborate on ideas about the data & coded categories, explores hunches
|
|
Why use grounded theory research?
|
1. to generate a theory
2. to explain a process, action or interaction 3. when you want a step-by-step systematic procedire 4. when you want to stay close to the data |
|
Why is qualitative research called "interpretive" research?
|
It involves making a personal assessment of descriptions and themes.
|
|
Agar (1980) recommends to "read the transcripts in their entirety several times. Immerse yourself in the details, trying to get a sense of the interview as a whole before breaking it into parts." This is a suggestion to engage in:
|
preliminary exploratory analysis.
|
|
A researcher who is segmenting and labeling the text is engaged in:
|
coding
|
|
Sentences or paragraphs that all relate to a single code are called:
|
test segments
|
|
Similar codes aggregated together to form a major idea in the database are called:
|
themes
|
|
In conducting qualitative research on student workload, a researcher will analyze views from students, student life personnel, university professors, and an academic administrator. The purpose of doing this is to develop:
|
multiple perspectives
|
|
Information that disconfirms the themes produced in the study is known as:
|
contrary evidence
|
|
Saturation in qualitative research refers to:
|
all themes have been identified
|
|
Generating a chronology of events or a conceptual model of events is known as:
|
interconnecting the themes
|
|
A written passage summarizing the findings of the data analysis is known in qualitative research as:
|
narrative discussion
|
|
Qualitative researchers include personal reflection on the meaning of the data:
|
because their views influence their interpretations
|
|
The process of corroborating findings from different individuals, data sources, or collection methods is known as:
|
triangulation
|
|
A third-party evaluation of different aspects of the research is known as
|
external audit
|