Summary Statistics for Measurement Variables by using Stata

Dataset electricity.dta contains information on electricity consumption in U.S. states, from the California Energy Commission (2012). To find the mean and standard deviation of per capita electricity use (elcap), type This table also gives the number of nonmissing observations and the variable’s minimum and maximum values. If we had simply typed summarize with no

Exploratory Data Analysis by using Stata

Statistician John Tukey assembled a toolkit of old and new methods for exploratory data analysis (EDA), which involves analyzing data in an exploratory and skeptical way without making unneeded assumptions (see Tukey 1977; also Hoaglin, Mosteller and Tukey 1983, 1985). Box plots, introduced in Chapter 3, are one the most popular EDA methods. Another

Normality Tests and Transformations by using Stata

Many statistical procedures work best when applied to variables that follow normal distributions. The preceding section described exploratory methods to check for approximate normality, extending the graphical tools (histograms, box plots, symmetry plots and quantile-normal plots) presented in Chapter 3. A skewness-kurtosis test, making use of the skewness and kurtosis statistics shown by summarize,

Frequency Tables and Two-Way Cross-Tabulations by using Stata

The summary statistics, graphs and transformations described above apply mainly to measurement variables. Categorical variables require different approaches, often starting with simple one- or two-way tables. For examples using such tables we return to the Granite State Poll data, Granite2011_6.dta. One question (trackus) asked whether people think the U.S. is headed in the right

1 Comments

Multiple Tables and Multi-Way Cross-Tabulations by using Stata

With surveys and other large datasets, we sometimes need frequency distributions of many different variables. Instead of asking for each table separately, for example by typing tabulate tparty, then tabulate obama, and finally tabulate trackus, we could simply use another specialized command, tabl: . tabl tparty obama trackus Or, to produce one-way frequency tables

Tables of Means, Medians and Other Summary Statistics by using Stata

Summary Statistics tabulate produces tables of means and standard deviations within categories of the tabulated variable. For the remaining examples in this chapter, we return to the data on electricity consumption in U.S. states. tabulate gives us one way to view summary statistics of per capita electricity consumption (elcap) for each of the 9

Using Frequency Weights with Stata

summarize, tabulate, table and related commands can be used with frequency weights that indicate the number of replicated observations. For example, here are the mean and other statistics for per capita electricity use across all U.S. states. This mean, 13,318 kWh, tells us the average electricity across the 51 states (including District of Columbia)

One-Sample Tests by using Stata

One-sample t tests have two seemingly different applications: Testing whether a sample mean y differs significantly from an hypothesized value p0 . Testing whether the means of yj and y2 , two variables measured over the same set of observations, differ significantly from each other. This is equivalent to testing whether the mean of

Two-Sample Tests by using Stata

The remainder of this chapter draws examples from a survey of college undergraduates by Ward and Ault (1990). About 19% of these students belong to a fraternity or sorority. On campus, these organizations and their members are commonly referred to as “Greeks” not with any reference to nationality, but because most of the organizations

1 Comments

One-Way Analysis of Variance (ANOVA) by using Stata

Analysis of variance (ANOVA) provides another way, more general than t tests, to test for differences among means. The simplest case, one-way ANOVA, tests whether the means of y differ across categories of x. One-way ANOVA can be performed by a oneway command with the general form oneway measurement categorical. For example, The tabulate

Two- and N-Way Analysis of Variance by using Stata

One-way ANOVA examines how the means of measurement variable y vary across categories of one other variable x. N-way ANOVA generalizes this approach to deal with two or more categorical x variables. For example, we might consider how drinking behavior varies not only by fraternity or sorority membership, but also by gender. We start

Factor Variables and Analysis of Covariance (ANCOVA) by using Stata

anova and many other Stata estimation commands allow independent variables specified in factor variable notation. The prefix i. written before the name of an independent variable tells Stata to include indicator (binary) variables for levels of a categorical variable, as if each category comprised its own dichotomous predictor. Categorical variables marked by the i.

Predicted Values and Error-Bar Charts by using Stata

After anova , the followup command predict calculates predicted values, residuals or standard errors and diagnostic statistics. One use for such statistics is in drawing graphical representations of the model results, such as an error-bar chart. For a simple illustration, we return to the one­way ANOVA of drink by year: . anova drink year

Simple Regression by using Stata

File Nations2.dta contains U.N. human-development indicators for 194 countries: . use C:\data\Nations2.dta, clear Life expectancies (life) exhibit much place-to-place variation. For example, Figure 7.1 shows that they tend to be lower in Africa than elsewhere. To what extent can variations in life expectancy be explained by average education, per capita wealth and other development

Correlation in Linear Regression by using Stata

Ordinary least-squares (OLS) regression finds the best-fitting straight line. A Pearson product- moment correlation coefficient describes how well the best-fitting line fits. correlate obtains correlations for listed variables. . correlate gdp school adfert chldmort life correlate reports correlations based only on those observations that have non-missing values on all of the listed variables. From

Multiple Regression by using Stata

Simple regression and correlation establish that a country’s life expectancy is related to the mean years of schooling: school by itself explains about 52% of the variance of life. But might that relationship be spurious, occurring just because both schooling and life expectancy reflect a country’s economic wealth? Does schooling matter once we control

1 Comments

Hypothesis Tests with Linear Regression by using Stata

Two types ofhypothesis tests appear in regress output tables. As with other common hypothesis tests, they begin from the assumption that observations in the sample at hand were drawn randomly and independently from an infinitely large population. Overall F test: The F statistic at the upper right in the regression table evaluates the null

1 Comments

Dummy Variables in Linear Regression by using Stata

Categorical variables can become predictors in a regression when they are expressed as one or more {0,1} dichotomies called dummy variables. For example, we have already seen large regional differences in life expectancies (Figure 7.1). The categorical variable region takes values from 1 (Africa) to 5 (Oceania) which can be re-expressed as a set

Interaction Effects in Linear Regression by using Stata

The previous section described what are called “intercept dummy variables,” because their coefficients amount to shifts in a regression equation’s y intercept, comparing the 0 and 1 groups. Another use for dummy variables is to form interaction terms called “slope dummy variables” by multiplying a dummy times a measurement variable. In this section we

Robust Estimates of Variance in Linear Regression by using Stata

The standard errors and hypothesis tests that accompany ordinary regression (such as regress or anova) assume that errors follow independent and identical distributions. If this assumption is untrue, those standard errors probably will understate the true sample-to-sample variation, and yield unrealistically narrow confidence intervals or too-low test probabilities. To cope with a common problem