ch 10

Compare and contrast the five measures of association.

1. Lambda- nominal, based on a reduction in errors based on the mode and ranges between 0 (independance) and 1.0 (perfect prediction or the strongest possible relationship). 2. Gamma- Ordinal, based on comparing pairs of variable categories and seeing whether a case has the same rank on each. Ranges from -1.0 to +1.0, with 0 meaning no association. 3. Tau- Ordinal, based on a different approach than gamma and takes care of a few problems that can occur with gamma. This is specifically Kendall's tau. It ranges from -1.0 to +1.0, with 0 meaning no association. 4. Rho, measures correlation, interval and ratio data is measured only. It ranges from -1.0 to +1.0, with 0 meaning no association. 5. Chi-squared- it is observing nominal and ordinal data for measures of association. Its upper limit is infinity and the lower limit is 0, which means no association.

Standard deviation-

A measure of dispersion for one variable that indicates an average distance between the scores and the mean.

Percentile-

A measure of dispersion for one variable that indicates the percentage of cases at or below a score or point.

Suppressor variable pattern-

A pattern in the elaboration paradigm in which no relationship appears in a bivariate contingency table, but the partials show a relationship between the variables.

Interpretation pattern-

A pattern in the elaboration paradigm in which the bivariate contingency table shows a relationship, but the partials show no relationship and the control variable is intervening in the casual explanation.

Specification pattern-

A pattern in the elaboration paradigm in which the bivariate contingency table shows a relationship. One of the partial tables shows the relationship, but other tables do not.

Replication pattern-

A pattern in the elaboration paradigm in which the partials show the same relationship as in a bivariate contingency table of the independent and dependent variable alone.

Regarding the elaboration paradigm, describe the five patterns that are possible when comparing bivariate tables with partial tables.

1. Replication Pattern- when partials replicate or reproduce the same relationship that existed in the bivariate table before you considered the control variable. It means that the control variable has no effect. 2. Specification Pattern- when onepartial replicated the initial bivariate relationship but other partials do not. Note- interpretation and explanation patterns both show a relationship in the bivariate table that disappear into the partials. 3. Interpretation pattern- describes the situation in which the control variable intervenes between the origional independent and dependant variables. It helps interpret the data. 4. Explanation pattern- looks at the same interpretation but the control variable comes before the independent variable in the initial bivariate relationship. 5. Suppressor variable pattern- occurs when the bivariate tables suggest independence when in reality the relationships appear in one or both of the partials.

Define, compare, and contrast statistical significance, level of significance, Type I errors, and Type II errors.

1. Statistical significance- A way to discuss the likelihood that a finding or statistical relationship in a sample is due to the random factors rather than due to the existence of an actual relationship in the entire population. Level of statistical significance- A set of numbers researchers use as a simple way to measure the degree to which a statistical relationship results from random factors rather than the existence of a true relationship among variables. 2. Type I error- The logical error of falsely rejecting the null hypothesis. 3. Type II error- The logical error of falsely accepting the null hypothesis. - Obviously errors are bad, but it is important to note that it would be better to falsely reject something than falsely accept something that will cause subsequent damage. Depending on how significant something is the more important/ statistically important the numbers are. And the statistical significance is important because it shows that the data is both important and gives it numerical backing.

Normal distribution-

A "bell-shaped" frequency polygon for a distribution of cases, with a peak in the center and identical curving slopes on either side of the center. It is the distribution of many naturally occurring phenomena and os a nasis of much statistical theory.

Control variable-

A 'third' variable that shows whether a bivariate relationship holds up to alternative explanations. It can occur before or between other variables.

Scattergram-

A diagram to display the statistical relationship between two variables based on plotting each case's values for both of the variables.

Pie chart-

A display of numerical information on one variable that divides a circle into fractions by lines representing the proportion of cases in the variables attributes.

Skewed distribution-

A distribution of cases among the categories of a variable that is not normal (not a bell shape) Instead of an equal number of cases on both ends, more are at one of the extremes.

Codebook-

A document that describes the procedure for coding variables and their location in a format for computers.

Descriptive statistics-

A general type of simple statistics used by researchers to describe basic patterns in the data.

Median-

A measure of central tendency for one variable indicating the point or score at which half the cases are higher and half are lower.

Mean-

A measure of central tendency for one variable that indicates the arithmetic average (i.e. the sum of all scores divided by the total number of scores).

Mode-

A measure of central tendency for one variable that indicates the most frequent or common score.

Range-

A measure of dispersion for one variable indicating the highest and lowest scores.

Curvilinear relationship-

A relationship between two variables such as the values of one variable increase, the values of the second show a changing pattern (e.g. first decrease then increase) It is not a linear relationship.

Level of statistical significance-

A set of numbers researchers use as a simple way to measure the degree to which a statistical relationship results from random factors rather than the existence of a true relationship among variables.

Elaboration paradigm-

A system for describing patterns evident among tables when a bivariate contingency table is compared with partials after the control variable has been added.

Frequency distribution-

A table that shows the distribution of cases into the categories of one variable (i.e. the number or percent of cases in each category). Independent- the absence of a statistical relationship between two variables. There is no association between them.

Statistical significance-

A way to discuss the likelihood that a finding or statistical relationship in a sample is due to the random factors rather than due to the existence of an actual relationship in the entire population.

Z-score-

A way to locate a score in a distribution of scores by determining the number of standard deviations it is above or below the mean or arithmetic average.

Linear relationship-

An association between two variables that is positive or negative across the attributes or levels of the variables. When plotted in a scattergram, when the basic pattern of the association forms a straight line, not a curve or other pattern.

What are control variables and why are they important?

Control variable- A 'third' variable that shows whether a bivariate relationship holds up to alternative explanations. It can occur before or between other variables. Control variables are very important because they provide an opportunity for essentially proving how variable relate and affect the results of the study.

Partials-

In contingency tables for three variables, tables that show the saaociation between the independent and dependant variables for each category of a control variable.

What is multiple regression, why is it used, and what information does it give?

It is a statistical technique whose calculation is often done through computer software. It creates an opportunity for control variables and controls alternative explanations. It also is widley used in social science, so it is helpful to understand and apply.

Code sheets-

Paper with a printed grid on which a researcher records info so that it can be easily entered into a computer. It is an alternative to the direct-entry method and uses optical-scan sheets.

Cross-tabulation-

Placing data for two variables in a contingency table to show the number or percentage of cases at the intersection of categories of the two variables.

Compare and contrast the three measures of central tendency.

Range- A measure of dispersion for one variable indicating the highest and lowest scores. Percentile- A measure of dispersion for one variable that indicates the percentage of cases at or below a score or point. Standard deviation- A measure of dispersion for one variable that indicates an average distance between the scores and the mean. Range is the simplest and it consists of the largest and smallest scores. Percentile tells the scores of specific places and standard deviation is the most commonly used.

Compare and contrast the three techniques that researchers use to decide whether a relationship exists between two variables.

Scattergram- A diagram to display the statistical relationship between two variables based on plotting each case's values for both of the variables. Cross-tabulation- Placing data for two variables in a contingency table to show the number or percentage of cases at the intersection of categories of the two variables. Measures of association/statistical measures- express the degree of covariation in a single number. Scatterplots provide a visible image to disern the information. Cross tabulation views percentages between two variables and a Measure of association is a single number that expresses the strength and often the direction of a relationship.

Univariate statistics-

Statistical measures that deal with one variable only.

Bivariate statistics-

Statistical measures that involve two variables only.

Covariation-

The idea that two variables vary together, such that knowing the values in one variable provides info about values found in another variable.

Type II error-

The logical error of falsely accepting the null hypothesis.

Type I error-

The logical error of falsely rejecting the null hypothesis.

Marginal-

The totals in a contingency table, outside the body of a table.

Bar chart-

a display of quantitative data for one variable in the form of rectangles where longer rectangles indicate more cases in a variable category. Usually, it is used with discrete data and there is a small space between rectangles. They can have a horizontal or vertical orientation. Also called bar graphs.

Explanation pattern-

a pattern in the elaboration paradigm in which the bivariate contingency table shows a relationship, but the partials show no relationship and the control variable occurs prior to the independent variable.