Reading...
Play button
Play button
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;
56 Cards in this Set
- Front
- Back
|
Two types of statistics
|
descriptive- describes something
inferential - to infer something |
|
|
Roles of research
|
to explore
to describe to evaluate to explain |
|
|
purpose of research
|
to define something, to understand the relationship
|
|
|
formula
|
trying to convert one set of characteristics into other characteristics
|
|
|
linkage
|
trying to determine how good that formula is
|
|
|
validity
|
when our statement or conclusions about empirical reality are correct
|
|
|
2 types of validity
|
causal validity
measurement validity |
|
|
causal validity
|
one is associated with the other
|
|
|
measurement validity
|
measuring what you say you're going to measure
|
|
|
Types of causal validity and what they mean
|
Internal validity – there are no external influences, two factors are solely related
External validity – the relationship can be seen in other places at other times Statistical conclusion validity- a change in one factor is statistically related to the change in another factor |
|
|
Types of measurement validity and what they mean
|
Face validity – The test is said to have face validity if it "looks like" it is going to measure what it is supposed to measure.
Content validity – does the measure match the degree of measurement caught (refers to the extent to which a measure represents all facets of a given social construct) Convergent validity – how is your measurement in relation to those already accepted (the degree to which an operation is similar to (converges on) other operations that it theoretically should also be similar to) Construct validity- How does one measurement relate to another measurement in terms of fear (refers to whether a scale measures or correlates with the theorized psychological construct that it purports to measure) |
|
|
Reliability
|
what you measure will be the same over and over again
|
|
|
Types of data
|
(FOIDOCS)
Focus group, Official records, Interview, Data analysis (primary and secondary analysis), Observation (complete participation, observer as participant, complete observer), Content analysis, Survey |
|
|
Unit of analysis
|
what you are studying, what is your subject/observation (ie. county, individual, case, year)
|
|
|
Population vs sample
|
Sample is a subset of a population
|
|
|
Probability sampling
|
everyone in the population is selected by a known probability
|
|
|
Types of probability sampling and their meaning
|
a. Simple random – each person has an equal chance of being selected
b. Systematic random sampling – the first person is chosen by simple random, then after that, everyone else is every nth selection. c. Multistage cluster- multiple stages to get your sample (out of four neighborhoods, you randomly select two of them, then you select five random blocks from those two chosen neighborhoods). You can do as many levels as you want. i. Stratified – separate the population into groups, then the next stages are random sampling ii. Weighted – put more weight in a naturally underrepresented sample or weight the data itself (like taking a score for women only and multiplying it by 3 times so it is closer to the sample of men). |
|
|
Non-probability sampling
|
everyone in the population does not have a chance
|
|
|
Types of non-probability sampling and their meaning
|
a. Purposive (snowballing)
b. Quota samples – numbers are predetermined (mall survey) c. Availability – convenient, whom ever comes across your way |
|
|
Variable
|
anything that varies
|
|
|
Constant
|
does not alter
|
|
|
Dependent variable
|
is just there. The variable that a researcher wishes to explain or predict
|
|
|
Independent variable
|
depends on the dependent variable. Variable that will explain or predict the dependent variable.
|
|
|
Empirical research
|
is real and observable
|
|
|
Qualitative
|
categorical (alphanumeric), examples include gender, geographic area, political party
|
|
|
Quantitative
|
continuous (numbers), examples include age, income, number of arrests
|
|
|
Four levels of measurement and their meanings
|
Qualitative/Categorical
Nominal – categorical in nature, groups without order (gender, race, religion) Ordinal – categorical in some type of order, rank ordered but can’t be quantified (Likert scale, or categories in ranges like income or age) Quantitative/Continuous Interval – equal and known quantity in a ranking order and zero is arbitrary (0 Fahrenheit does not mean that there is no temperature, it just means it is colder) Ratio – same as interval but does have a true zero point (income) |
|
|
3 types of research design
|
pre-experimental
experimental quasi-experimental |
|
|
pre-experimental design types
|
One-shot case study (most social science studies are this) XO
One-group pretest-post OXO Static group comparison (comparing results of someone who has had the treatment and someone who has not) XO O |
|
|
experimental design types
|
Pretest-posttest control group (treatment group gets a sugar pill) ROXO
RO O Solmon four-group (different types of observation and pre/post test with each group) ROXO RO O R X O R O Post-test control group (comparing results of someone who has had the treatment and someone who has not) R XO R O |
|
|
Quasi-Experimental design types
|
Untreated control group w/ Pretest-posttest (same as Pretest-posttest control group, but not randomly assigned)
OXO O O Time series (multiple data points before and after treatment) OOOOXOOOO Time series w/ control group (multiple data points before and after treatment for one group, then multiple data points without treatment for another group) OOOOXOOOO OOOO OOOO Longitudinal OOOOOOOOOOO |
|
|
Internal validity
|
Did the treatment make a difference in this instance?
|
|
|
Types of internal validity and their meaning
|
• History: the specific events which occur between the first and second measurement.
• Maturation: the processes within subjects which act as a function of the passage of time • Testing: the effects of taking a test on the outcomes of taking a second test. • Instrumentation: the changes in the instrument, observers, or scorers which may produce changes in outcomes. • Statistical regression: This threat is caused by the selection of subjects on the basis of extreme scores or characteristics • Selection: the biases which may result in selection of comparison groups. Randomization (Random assignment) of group membership is a counter-attack against this threat. However, when the sample size is small, randomization may lead to Simpson Paradox, which has been discussed in an earlier lesson. |
|
|
External validity
|
Generalizability to other similar situations
|
|
|
Types of External validity and their meaning
|
• Reactive or interaction effect of testing: a pretest might increase or decrease a subject's sensitivity or responsiveness to the experimental variable. Indeed, the effect of pretest to subsequent tests has been empirically substantiated (Wilson & Putnam, 1982, Lana, 1959).
• Interaction effects of selection biases and the experimental variable • Reactive effects of experimental arrangements: it is difficult to generalize to non-experimental settings if the effect was attributable to the experimental arrangement of the research. • Multiple treatment interference: as multiple treatments are given to the same subjects, it is difficult to control for the effects of prior treatments. |
|
|
types of data representations
|
Bar charts
Histograms Pie chart |
|
|
Courtoisis
|
is the peakness of the distribution
|
|
|
Platykurtic
|
way above the expected curve (a huge spread of data)
|
|
|
Leptokurtic
|
below (practically everyone is the same, like the Rocketts)
|
|
|
What is the difference between negative and positive skewness?
|
Negative skew has a tail to the left and positive skew is to the right.
|
|
|
What causes skewness?
|
Usually caused by an outlier in the data
|
|
|
Central tendency
|
captures the most typical score in a distribution of scores
|
|
|
What are the types of central tendency and which level of measurement are they used with?
|
Mean - continuous
Median - continuous Mode - categorical |
|
|
How do you calculate Mean?
|
total all of the values and divide them by the number of values
|
|
|
Hows do you calculate Median?
|
Step 1 -rank order, Step- 2 find the position locator
Position locator = N+1 divided by 2, where N is how many scores/ranks there are Step 3- If N is not even, take average of number before and after position, add those two numbers and divide by 2 to get the median number. |
|
|
How do you calculate Mode?
|
what is the highest number
|
|
|
Measures of Dispersion
|
assesses the spread of the data
|
|
|
Heterogeneity
|
a lot of variability (like in platykurtic kurtosis)
|
|
|
Homogenous
|
very little variability (like in leptokurtic kurtosis)
|
|
|
What measure of dispersion is used for categorical data?
|
Variation ratio
|
|
|
What does Variation ratio do?
|
measures the extent to which observations are not in the modal category.
|
|
|
What is the formula for Variation ratio?
|
VR= 1- fmc/n
then multiply by 100 to get % |
|
|
What measure of dispersion is used for continuous data?
|
range
variance standard deviation |
|
|
What does the Standard deviation tell you?
|
distance between a score and the mean of the group scores (how spread the data is from the mean).
|
|
|
What is the formula for Standard Deviation?
|
s=
|
|
|
What are the reasons for each step of the Standard Deviation?
|
Step 1: subtract the mean from each score/taking the deviation from the mean.
Step 2: sum of deviation scores Step 3: Sum of squares to resolve the issue that step 2 equals zero Step 4: Divide by the number of scores minus 1 to average and -1 to adjust for sample bias Step 5: square root to get back to scale |
