Flashcards in Exam 3 Deck (96):
The process of specifying what is meant by a term. Its different aspects and different dimensions, how it is different from existing concepts.
The process of connecting concepts to observations.
The level at which the data is actually being collected. Collection of operation indicators. (answers to survery's, drawing blood samples)
Cover all possible responses- Can everyone in the study be classified or scored?
When every case can be classified as having only one attribute (value)-Can the answer only have one value?
Nominal level of measurement
Variables whose values have no mathematical interpretation; they vary in kind or quality but not in amount-Color, name, country, hometown.
Ordinal level of measurement
The numbers indicating a variable's values specify only the order of the cases, indicates the ranks.
Interval level of measurement
The numbers represent fixed measurement units but have no absolute zero point. Numbers can be added and subtracted, but ratios are not meaningful.
Ratio level of measurement
The numbers represent fixed measuring units and an absolute zero point (meaning no amount of whatever the variable indicates).-Income, household size, years married.
Singular view (looks at one thing), expert opinion (no data). Doesn't provide convincing evidence of measurement validity.
Expansive view, expert opinion. Often uses index set of questions.
Singular view, empirical testing(uses data). Compares stats to an established similar variable.
Validity exists when a measure yields scores that are closely related to scores on a criterion measured at the same time.
Validity is the ability of a measure to predict scores on a criterion measured in the future.
Expansive view, empirical testing. Established by showing that measures are related to a variety of other measures.
Validity is acheived when one measure of a concept is associated with different types of measures of the similar concepts.
Validity is a complementary approach to construct validation. Compared on scores of measures of different but related concepts.
The same measure at two different points in time. Examines the stability of the measure over time.
Examines the internal consistency of the measure.
The statistical test for inter-item reliability. Above .70 indicates good.
Examines the equivalence of measures in different forms. Ex. English-Spanish survey.
Do different observers report similarly? Examines whether there is agreement/consistency.
Allows us to know in advance how likely it is that any element of a population will be selected for the sample.
Simple Random Sampling
Identifies each element strictly on the basis of chance.
Systemic Random Sampling
First element is selected randomly and then every nth element is selected. Need to make sure there are no problems with peiodicty.
Stratified Random Sampling
All elements in the sampling are distinguished according to their value on some characteristics.
Multi-State Cluster Random Sampling
Useful when collecting comprehensive list, or when collecting data from a simple random sample.
Sampling methods that do not let us know in advance the likelihood of selecting each element.
Elements are selected for availabilty sampling because they are available or easy to find.
Selecting a pre-set number of elements based on characteristics in a population to ensure that the sample represents those characteristics in proportion to their prevalence in the population.
Each sample element is selected because of the unique and unusual position of the sample elements.
There is no sampling frame, but the members of which are somewhat interconnected.
Gathers data from every element of the entire target population. It is a way to get participants, but technically is not a "sample" since it includes everyone.
Entire set of individuals or other entities to which study findings are to be generalized
A list of all elements or other units containing the elements in a population
The individual members of the population whose characteristics are to be measured
Overrepresentation or underrepresentation of some population characteritics in a sample resulting from the method used to select the sample.
Does the sample come close to representing the target population?
Simple random sampling
A method of sampling in which every sample element is selected only on the basis of chance, through a random process.
Random number table
A table containing list of numbers that are ordered solely on the basis of chance, it is used for drawing a random sample
Random digit dialing
The random dialing by a machine of numbers within designated phone prefixes, which creates a random sample for phone surveys.
The number of cases from one sampled case to another in a systematic random sample.
A sequence of elements that varies in some regular, periodic pattern
Stratified random sampling
A method of sampling in which sample elements are selected separately from population strata that are identified in advance by the researcher. Division of a population into smaller groups.
Sampling method in which elements are selected from strata in exact proportion to their representation in the population.
Sampling in which elements are selected from strata in different proportions from those that apepar in the population.
A naturally occurring mixed aggregate of elements of the population.
Sampling in which elements are selected on the basis of convenience.
A non probability sampling method in which elements are selected to ensure that the sample represents certain characteristics in proportion to their prevalene in population.
Non probability sampling method in which elements are selected for a purpose, usually because of their unique position.
Where existing study subjects recruit future subjects from among their acquaintances.
Sample size increases the power of the study..
by shrinking the confidence interval, there is less chance of error. Small samples make an increase in the confidence interval.
Cross-Sectional research design
A study in which data are collected at only one point in time.
Longitudinal Research Design
A study in which data are collected that can be ordered in time; also defined as researchin which data are collected at tow or more points in time.
Goal: identify the trend of change in the population, not the individuals.
Panel-Fixed-Sample Panel Design
Same participants were followed up after the initial data collection.
Goal: Examine the change of some individuals or groups.
Data from the same cohort (those who share a common event or starting point) were collected at multiple time points.
Event based repeated cross sectional design
Different samples from same cohort.
Event based Panel design
Same individuals from same cohort are followed up over time.
Unit of analysis
The level of social life on which a research question is focused, such as individuals, groups, towns, or nations.
Unit of observation
Researchers base conclusions on information that is collected and analyzed, so using defined units of observation in a survey or other study helps to clarify the reasonable conclusions that can be drawn from the information collected.
An error in reasoning in which incorrect conclusions about individual-level processes are drawn from group-level data.
An error in reasoning that occurs when incorrect conclusions about group-level processes are based on an individual.
Idiographic Causal Explanations
Identifies the concrete individual sequence of events, thoughts, or actions that resulted in a particular outcome for a particular individual.
Nomothetic Casual Explanations
Identifies common influences on a number of cases or events. Typically, the independent variable is the presumed cause and the dependent variable is the potential effect.
5 Criteria for nomothetical casual explanations.
Relationship between two variables.
Association in experimental research
Determine association by manipulating the independent variable.
Association in non-experimental research
Determine association based on whether values of cases that differ on the independent variables tend to differ in terms on the dependent variable.
The variation in the dependent variable occurred after the variation in the independent variable
Cross sectional design
Data is collected at a single time point.
Data is collected at more than one time point
Time order in experimental research
Controlled by researchers
Time order in non-experimental research
Our ability to establish time order depends on the timing of measurement
A criterion for establishing a causal relation between two variables; when a relationship between two variables is not caused by variation in a third variable.
Non-spuriousness in experimental research
Randomization--The use of random techniques to designate subjects into comparison groups
Non-spuriousness in non-experimental research
Statistical control-- A method in which the effects of hugely influential variables are taken into account so that there can be more confidence on the effects of the predictors
A discernible process that creates a causal connection between variation in an independent variable and the variation in the dependent variable it is hypothesized to cause.
A set of interrelated circumstances that alters a relationships between other variables or social processes.
Variables that define context
5 Causal Criterial
Three types of longitudinal designs
The random assignment of cases as by the toss of a coin.
A method in which one variable is held constant so that the relationship between two other variables can be assessed without the incluence of variation in the control variable
A researcher is going to use a phonebook in Ames to investigate how many families with landline phones also carry cellphones.
A member of ISU diversity committee is attending the university lecture series searching for students representing diverse backgrounds to sit on panels and in focus groups
A researcher wants to do a representative study about immigration experiences comparing those who came to the US to seek jobs and those who came to escape the warfare in their home country.
A researcher is collecting love stories, from veterans of WWII, and begins by interviewing her grandfather and a prominent neighbor who happens to live down the street
a student researcher is going to investigate the quality of janitorial services at isu by taking dust samples from the floor space in 25 classrooms- the university however, doesn't want to furnish the student with a list of classrooms for this purpose. The results will be compared to a nationwide average
probability-- multi-stage cluster
a researcher wants to evaluate satisfaction with the concessions prices for season ticket holders of the Chicago cubs baseball team. The report will be used by executives to adjust prices at the stadium.
A student wants to gather a few quotes in support of or against his review of the newly-released Hollywood flow : voldemort vs. vader.
non-probably- convenience / availability
A researcher who is investigating the relationship quality of first-time mothers with friends and family members who provide social support during the birthing process seeks out and interviews a midwife who has helped deliver more than 1000 babies in homes during her career
non probability- purposive
Know the rule of thumb used in this class about sample size.
1,000 is good, 100 is okay, 10 is worthless.
The more participants, the better the accuracy.
All things being equal, relates to experimental designs and the use of statistical control variables.