Political Research Exam 2

Simple Random Sample (example)

a random selection number generator

Cumulative Frequency

adding up all of the previous values up to the current point

How can a z-score be interpreted in a normal distribution?

deviation of the group from the mean

Elements

individual units, unit of analysis

Frequency Distribution

number of times a variable occurs

Systematic Sample (example)

picking every fifth name in a list

For what purpose might you use a disproportionate sample?

when the proportionate sample of the group is too small to draw a conclusion about the population

Stratified Sample (example)

300/500 has a larger representation in the sample than 100/500.

Example of a Sample

A State/Region of the US

What is a normal distribution?

A distribution defined by a mathematical formula and the graph f which has a symmetrical bell shape in which the mean, the mode and the median.

Standard Deviation

A measure of dispersion of data points about the mean for internal and ratio-level data.

Convenience Sample (definition)

A nonprobability sample in which the selection of elements is determined by the researcher's convenience.

What is the difference between a one- and two-tailed test?

A one-tailed test indicates a direction, a two-tailed test does not indicate a direction.

Simple Random Sample (definition)

A probability sample in which each element has an equal chance of being selected.

Systematic Sample (defintion)

A probability sample in which elements are selected form a list at predetermined intervals.

Stratified Sample (definition)

A probability sample in which elements sharing one or more characteristics are grouped and elements are selected from each group in proportion to the group's representation in the total population.

Cluster Sample (definition)

A probability sample that is used when no list of elements exists. The sampling frame initially consists of clusters of elements.

What are open-ended questions? (advantages/disadvantages)

A question with no response alternatives provided for the respondent. Respond too much or too little.

What are closed-ended questions? (advantages/disadvantages)

A question with response alternatives provided. Easy to answer and takes little time to answer, usually more willing to answer personal questions, clarify questions Force respondent to choose an answer that may not accurately represent their view, provides oversimplified or extorted view of public opinion

Snowball Sample (definiton)

A sample in which respondents are asked to identify additional members of a population.

Alternative Hypothesis

A statement about the value or values of a population parameter. A hypothesis proposed as an alternative to the null hypothesis.

Null Hypothesis

A statement that a population parameter equals a single or specific value; the hypothesis that these is no relationship between two variables in the target population

Disproportionate Sample

A stratified sample in which elements sharing a characteristic are underrepresented or overrepresented in the sample.

Stratum (definition)

A subgroup of a population that shared one or more characteristics.

Sample

A subset of observations or cases drawn from a specified population.

Population

All the cases or observations covered by a hypothesis; all the units of analysis to which a hypothesis applies.

Cluster Sample (example)

Divide into clusters of same elements, then randomly choose clusters

What are common errors in designing survey questions?

Double barreled question: two questions in one Ambiguous question: what is your income? This is open-ended Leading question: asking a question that seems like the interviewer is looking for a specific response

What is a Type 1 error?

Error made by rejecting a null hypothesis when it is true.

What is the importance of random sampling as it relates to statistical inference? (Include: value, standard error, and sampling distribution in your answer)

Expected value: The average value of a sample statistic based on repeated samples. Expected to return the same value every time Standard error: provides the numerical indicator of the variation of the sample estimates. Discrepancy between an observed and true value. Statistical inference is the conclusions you draw from the study. It is important to have random sampling so it can be a true representation of the population

Why is it often necessary to sample a population?

Imagine the population you want to study is large. There's no good practical way to study most populations.

What are two reasons why there might be sampling bias?

It is usually due to an incomplete sampling frame or a non-probability method of selecting.

What is the difference between a longitudinal vs. cross-sectional survey?

Longitudinal: measures of variables at different points in time, instead of relying on previous works. The ability to measure the dependent variable multiple times Cross-sectional: A research design in which measurements of independent and dependent variables are taken at the same time; naturally occurring differences in the independent variable are used to create quasi-experimental and quasi-control groups; extraneous factors are controlled for by statistical means.

Mail, e- mail and Internet are low cost; why?

Low cost because they do not require interviewer time, but must be set up by a technical professional.

What is the trade-off in using mail, e-mail and internet in terms of completion rate?

Lower response rates

What are the 3 measures of central tendency? What are the advantages and disadvantages of each?

Mean (the average): appropriate for interval and ratio, truly quantitative measures, not good if there are extreme outliers that will skew the data Median (value divides the distribution in half): Isn't affected by outliers, applicable to ordinal, interval, and ratio data. Mode ( the data that appears the most): Works for nominal and categorical data, Message conveyed by one number is not representative of the entire data set

What is the difference between a critical and observed value?

Observed value is what you obtain from the test statistic to compare to the critical value. Critical value is the maximum distance from the mean you will accept for random error. Look up in table.

Relative Frequency

Percentage or proportion of total number of observations in a frequency distribution that have a particular value.

What is the alpha or level of significance?

Refers to the probability of making a type 1 error

Which is likely to have the highest completion rate? (Surveys or Interviews)

Surveys

Examples of Sampling Stratum

Taking a survey of students, separate by major, class, gpa

Example of a Population

The United States

Sample bias

The bias that occurs whenever some elements of a population are systematically excluded from a sample.

What is a z score and

The number of standard deviation by which a score deviates from the mean score.

Sampling frame

The population from which a sample is drawn. Ideally, it is the same as the total population of interest to study.

What is response rate and why does it matter?

The proportion of respondents selected for participation in a survey who actually participate. Low response means for a bad statistical inference, if people systematically do not participate, then it shows bias .

Confidence Interval

The range of values into which a population parameter is likely to fall for a given level of confidence.

Why do we usually use the relative frequency?

They put the raw frequencies into perspective

What does sampling have to do with statistical inference?

We base inferences ( in order to generalize) on a sample population.

What is the difference between a z and t distribution?

Z-tests compare population means to a sample's T-tests: statistically significant difference between two independent sample groups.