Research Methods Practice Quiz

_____________ is a form of qualitative research that focuses on one or more generally large or complex cases, such as organizations or communities.

Case study research

A _____________ relationship refers to how change in one variable produces or leads to change in another

Causal

A researcher looking to test whether two categorical variables are related, as in the form of a cross-tabulation, would want to conduct a:

Chi square

A ______________ helps researchers to understand how data were collected, coded, weighted, and so forth.

Codebook

_____________ guides the interviewer in asking questions and allows direct entry of the responses into an electronic database during a telephone interview.

Computer-assisted telephone interviewing

______ refers to clarifying what you want to measure through the measurement process, by defining it carefully and precisely.

Conceptualization

______ _______ tell the researcher the range of values in which the true parameter that is being estimated most likely falls, based upon what is learned from the sample.

Confidence attributes

A _______ is a statement of statistical sampling precision.

Confidence interval

______________ validity asks whether the measure includes all the important dimensions of the construct.

Content

__________ data contain measurements taken at a single point in time, such as student test scores from one school year.

Cross-sectional

______________ is a method used to describe relationships between nominal or ordinal categorical variables.

Cross-tabulation

A ______ is a pattern of values spread out over the categories or numeric range of a variable.

Distribution

If a researcher is interested in calculating a measure of how big a difference or effect is relative to the ordinary variation that the variable undergoes, then he/she should utilize the computation for ____.

Effect Size

A researcher has coded a set of dummy variables representing multiple categories so that the regression coefficients turn out to be contrasts with a grand mean. This method is known as:

Effect coding

__________ is an important tradition in qualitative research that come from anthropology and involves extended participant observation in a setting.

Ethnography

Through _________, the researcher begins by doing systematic observation of the world and then develops a logical explanation, versus ________________, where the researcher moves straight to the development of a logical explanation or theory and later gathers evidence to test the theory.

Induction/Deduction

______________ are the financial, human, and material resources required by a program, while ___________ are the immediate products of these activities.

Inputs/outputs

______ represent a mode of survey research where the interviewer stops people in public places to ask them questions.

Intercept interview surveys

______________ are large e-mail lists of respondents who opt in to participate in online surveys on various topics.

Internet access panels

In qualitative research, data are often spoken or written, or involve images with symbolic content. This makes __________ an important part of qualitative data analysis.

Interpretation

___________ data combines different levels, and is also known as hierarchical data.

Multilevel

In research, __________ predicts or explains a dependent variable using several independent variables.

Multiple regression

_________ was developed for some of the first major government and social surveys where interviewers went door to door to seek responses from participants.

Multistage sampling

Professor Smith is looking to test for differences between variables. She develops a ___ hypothesis which is a statement of no difference or no relationship, and a(n) ____ hypothesis, which indicates that there is a difference and a relationship.

Null, alternative

The _______________ is an estimate of how many subjects are needed to get the precision required for the study.

Sample size calculation

______ is a process of selecting people (or elements) for inclusion in a research study.

Sampling

_______ occurs when a sample gives systematically different results, because of shortcomings in the sampling process, from what would be obtained by studying the whole population:

Sampling bias

When a sampling procedure is repeated many times, drawing samples of the same size from which to make inferences of the population, a ________ is produced.

Sampling distribution

______________ uses a best-fit straight line to describe how one quantitative variable predicts another...

Simple regression

__________ refers to respondents providing socially acceptable answers, often due to the physical presence of an interviewer.

Social desirability bias

____ _____ are increasingly used as part of geographic information science to identify geographic coordinates and calculate distances.

Spatial data

The _____________ refers to how far the scores are, on average, from the mean.

Standard deviation

The ____ tells us the typical amount of sampling error we can expect in our point estimate.

Standard error

_________ refers to the typical error in the results of a sample when compared to the population.

Standard error

In preparing an effective questionnaire, a researcher should:

Start with the survey purpose and constructs Prepare mock tables and charts of the results Make sure to hook respondents with the first few questions

In research, drawing conclusions about an unknown population parameter based upon a sample estimate is referred to as:

Statistical Inference

Epistemology refers to:

The many ways of knowing, including use of the scientific method

Unit of analysis refers to:

The objects or things being described by the variables in a model

A correlation is a measure of the strength and direction of a relationship between two quantitative variables.

True

A limited dependent variable is one that cannot take on all possible values (such as negative values).

True

A sampling frame is an operational representation of the population from which the sample can be selected.

True

An example of protecting privacy and ensuring that information / data are nonidentifiable is the Health Insurance Portability and Accountability Act (HIPAA).

True

Analysis of variance (ANOVA) is a statistical method derived from experimental research which compares the means of the dependent variable across treatment categories.

True

Categorical variables may have more than two possible values.

True

Establishment surveys measure the traits of individuals mainly through in-person interviews in their homes.

True

Hypothesis are a prediction of what will happen if a theory is correct:

True

Intersubjectivity is the ability to understand the world from someone else's perspective.

True

Intervening variables represent steps in the causal process leading from the independent to the dependent variable

True

Logic models describe how programs or interventions produce desired outcomes.

True

Mail self-administered surveys are a mainstay of government statistical agencies, including the U.S. Census.

True

Measurement begins with a construct (trait), or the concept or thing that we are looking to measure.

True

Measurement involves protocols, or procedures for using instruments properly.

True

Misleading measurements, samples, and correlations are all ways that evidence can be wrong and embarrassing.

True

Online data analysis tools allow users to directly analyze data on the web, and have become an increasingly popular way for government agencies and other organizations to make their data publicly available.

True

Qualitative research allows us to better understand cultures or subcultures, whose communication modes may be expressed in different languages and other forms of symbolic systems.

True

Random sampling uses chance to select people from a population.

True

Since we live in an "information age," the ability to make use of complex sources of information is becoming an increasingly important skill.

True

Stratified sampling can be used to oversample selected strata and examine subgroups separately.

True

The p value is the probability associated with a test statistic, and represents the probability of observing the sample estimate when the null hypothesis about the population is true.

True

The scientific method originated with the natural sciences

True

The variance is the square of the standard deviation, and refers to the general idea of variability in data.

True

Unstructured interviews have no predetermined set of questions, and may seem like an ordinary conversation even though they are different.

True

When considering whether to do a survey, researchers should consider:

Whether enough is known about the topic to ask specific, structured questions Whether the information already exists in another source Whether respondents will provide truthful answers

_________ refers to the consistency of a measure and is directly related to the concept of random error.

Reliability

A researcher is interested in whether a variable really adds predictive power. Using the ________ adjusts for the number of independent variables in the multiple regression equation.

Adjusted R-Squared

Research methods:

Are techniques that produce evidence Involve sampling, measurement instruments, planned comparisons, and statistical techniques Allow us to judge the quality of the study and the evidence presented

When wording the questions of a survey, researchers should make sure to:

Avoid a level of specificity that exceeds the respondent's potential for having an accurate, ready-made answer Avoid bias from unequal comparisons Choose question wordings that allow essential comparisons to be made with previously collected data

Researchers use ________ to group many variables or indicators into a smaller set of underlying factors.

Factor analysis

Confidence intervals always reflect all sources of error.

False

Remote databases are composed of various tables of information that are linked and work together. These are common in maintaining administrative records.

False

Research is always perfect, and does not have any noticeable weaknesses.

False

Systematic measurement errors are those errors that are random and average out to zero.

False

The mean is always the best measure of center since it captures the full range of scores.

False

A researcher is interested in determining how study habits differ for male versus female students. She gives a variable a value of 1 if the student is female, and 0 if the student is male. The researcher is using a:

Female dummy variable

______________ are one of the most widely used methods of qualitative research, consisting of a small group of individuals who are gathered together and asked questions about a topic by a moderator:

Focus groups

________ uses data from a time series to predict future values of the dependent variables.

Forecasting

___________ is the ability to project the results of one study to a much larger reality.

Generalizability

An advantage of qualitative research is:

Generating exploratory studies of new or vaguely understood behaviors Better understanding the nature of group dynamics Analyzing historical or archival texts

Sources for theories include:

Grand social theories Induction and deduction Exploratory and Qualitative research

Validity is:

How well the measure actually represents the true construct of the thing being measured

Theories help to:

Identify key variables Tell causal stories Explain variation

A necessary step in the survey research process involves:

Identifying the population and sampling strategy Developing a questionnaire and survey procedures Collecting data

A researcher may use a ________ for doing multiple regression with a dummy variable as the dependent variable, where the coefficients can be interpreted as the predicted probability that the dependent variable will occur.

Linear probability model

______ data are collected over time, and may compare findings from point to point for individuals or events.

Longitudinal

______ refers to the process of systematically observing some feature or characteristic of the world and then recording it as a number or category.

Measurement

_____________ refers to "data about data," as a way of organizing and searching archives.

Metadata

The term ______ refers to data at its most basic level of observation.

Microdata

A survey approach which uses more than one type of data collection method is referred to as:

Mixed mode survey

A _______ is a graphical or mathematical representation that helps to articulate and communicate a theory

Model

_____________ refers to repeating a study with a different sample in a different place, time, or context using a different study design.

Replication

_____________ refers to the specific procedure a researcher carries out in making and recording a measurement.

Operationalization

Primary research can involve:

Original collection of data Original analysis of other researchers' data Original exploration of existing surveys, records, or transcripts

Surveys where the same respondents are tracked and repeatedly surveyed over time are known as:

Panel surveys

The level of participation-observation in a qualitative study includes:

Participant as observer Observer as participant Complete observer

Professor Smith wants to explore the relationship between study habits, socializing, and going to class. She decides to use _______ to help estimate the pattern of relationships between variables in a presumed causal structure.

Path analysis

An agency analyst finds that her independent variable is a perfect linear combination of two more of the other independent variables. This is referred to as:

Perfect multicollinearity

Research is an important aspect of modern public policy, as evidenced by applications in:

Performance measurement Evaluation research Evidence-based policy and programs

Statistical inference is comprised of:

Point estimates Precision Significance tests

A __________ relationship means that the independent and dependent variable move in the same direction, such that if one goes up or down, so does the other

Positive

In health care studies, ___________ refers to the number or share of a population that has a condition or disease, while ____________ is the rate at which new cases of a disease appear in a population.

Prevalence, Incidence

A ______ is based upon randomly choosing participants from a population for study.

Probability sample

__________ deals with the measurement of latent traits or constructs by using composite measures made up of multiple items.

Psychometrics

SSecondary data may not always provide an effective way to answer a research question. Researchers may find that they must collect their own primary data because:

Public sources do not provide enough data for small area studies such as neighborhoods or cities Exisiting data may not have the right combination of variables Existing data maybe outdated

Qualitative research often involves:

Purpose or theoretical sampling A small n (number of cases/participants) Choosing people or cases for a specific purpose or to generate theory

______________ sampling starts by dividing the population into groups, such as male or female.

Quota

Statisticians in the 1960s developed a method of sampling that gives both listed and unlisted telephone numbers an equal chance of being selected. This is known as:

Random Digit Dialing

The term ______ refers to how rapidly a social or economic condition changes over time.

Rate of change

An example of a data archive would be:

Survey documentation and analysis (SDA) Council of European Social Science Data Archives (CESSDA) Inter-University Consortium for Political and Social Research (ICPSR)

The ______ emphasizes the importance of all components of the mail survey, and stresses the need to adapt the approach to each situation.

Tailored design method

Generalizability refers to:

The ability to take the results of research and apply them to other situations other than the exact one that was carried out