Using Weights in Stata

Stata understands four types of weighting: aweight   Analytical weights, used in weighted least squares (WLS) regression and similar procedures. fweight    Frequency weights, counting the number of duplicated observations. Frequency weights must be integers. iweight    Importance weights, however you define importance. pweight   Probability or sampling weights, proportional to the inverse of the probability

Creating Random Data and Random Samples in Stata

The pseudo-random number function runiform() lies at the heart of Stata’s ability to generate random data or to sample randomly from the data at hand. The Base Reference Manual (Functions) provides a technical description of this 32-bit pseudo-random generator. If we presently have data in memory, then a command such as the following creates

Writing Programs for Data Management in Stata

Data management on larger projects often involves repetitive or error-prone tasks that are best handled by writing specialized Stata programs. Advanced programming can become very technical, but we could also begin by writing simple programs that consist of nothing more than a sequence of Stata commands, typed and saved as a text file. Text

Histograms by using Stata

Histograms, displaying the distribution of measurement variables, are most easily produced with their own command histogram. For examples, we return to the data on 194 nations seen earlier in Chapter 2, containing human-development indicators gathered by the United Nations. Figure 3.1 shows a simple histogram of adfert, the adolescent fertility rate. It was produced

Box Plots by using Stata

Box plots convey information about center, spread, symmetry and outliers at a glance. For example, Figure 3.4 is a simple box plot of adfert (adolescent fertility rate) obtained by typing Figure 3.4 confirms the positive skew of this distribution, and shows five high outliers. The box in a box plot extends from approximate first

Scatterplots and Overlays by using Stata

Scatterplots belong to a broad family called twoway graphs. Stata’s basic scatterplot command has the form . graph twoway scatter y x where y is the vertical or y-axis variable, and x is the horizontal or x-axis one. (The initial graph twoway part of this command is optional, but kept here to emphasize a

Line Plots and Connected-Line Plots by using Stata

Mechanically, connected-line plots (graph twoway connect) are just scatterplots in which the points are connected by line segments. Line plots (graph twoway line) show the line segments without markers for the scatterplot points. Both belong to Stata’s versatile graph twoway family, which can be overlaid in any combinations. The scatterplot options that control axis

Other Twoway Plot Types by using Stata

In addition to basic line plots and scatterplots, the graph twoway command can draw a wide variety of other types. This section illustrates several more; type help graph twoway for a complete list. Earlier, in Figures 3.10 and 3.11, we used graph twoway lfit (linear fit) to draw a simple regression line. A similar

Bar Charts and Pie Charts by using Stata

The graph bar command, unlike graph twoway bar, works well to display relationships involving one or more categorical variables. Such graphs prove particularly useful with survey data, as will be shown in Chapter 4. This section serves just to introduce the command, with an example using variables from the cross-national dataset Nations_2.dta. . use

Symmetry and Quantile Plots by using Stata

Box plots, bar charts and histograms summarize measurement variable distributions, hiding individual data points to clarify overall patterns. Symmetry and quantile plots, on the other hand, include points for every observation. They take more effort to read than summary graphs, because they convey more detailed information. A histogram of the ratio of females to

Adding Text to Graphs by using Stata

Titles, captions and notes can be added to make graphs more self-explanatory. The default versions of titles and subtitles appear above the data region; notes (which might document the data source, for instance) and captions appear below. See Figure 3.7 for an example using title, caption and note. These defaults can be overridden, of

Graphing with Do-Files by using Stata

Complicated graphics like Figure 3.26 require graph commands that are many physical lines long (although Stata views the whole command as one logical line). Do-files, introduced in Chapter 2, help in writing such multi-line commands. They also make it easy to save the command for future re-use, in case we later want to modify

Retrieving and Combining Graphs by using Stata

Any graph saved in Stata’s “live” .gph format can subsequently be retrieved into memory by the graph use command. For example, we could retrieve Figure 3.26 by typing . graph use fig03_26.gph Once the graph is in memory, it is displayed onscreen and can be printed or saved again with a different name or

1 Comments

Graph Editor in Stata

The Graph Editor allows us to alter the appearance of a graph currently in memory, whether just drawn or previously saved and retrieved by graph use. It is easier to learn about this useful feature by experimenting yourself, rather than reading from a book. As an example to get started, however, we can show

Creative Graphing by using Stata

Edward Tufte, in his elegant and influential books about graphing data (1990, 1997, 2001, 2006), calls for more effort at designing clear, information-packed graphics. Presenting a rich collection of impressively good or humorously awful examples, Tufte shows how successful graphics allow viewers to draw their own comparisons and examine details of relationships between variables.

Declare Survey Data in Stata

Since 2001, the Granite State Poll at the University of New Hampshire has conducted statewide telephone surveys several times each year. Each survey contacts a new sample of about 500 people, asking a variety of opinion questions along with respondent background characteristics. The poll’s political findings attain national importance every four years during New

Design Weights of Survey Data in Stata

The previous section took weight definitions for granted, and indeed many data users begin with a completed survey in which weights have already been calculated by someone else. This and the following sections present examples showing how such calculations are done. Survey researchers apply probability weights to adjust for biases in their sampling methods.

Poststratification Weights of Survey Data in Stata

The previous section gave an example of weights based on the sampling design, which was known before data collection began. A second type of weights might be defined after we have collected the data, and see that despite our best efforts, it appears unrepresentative in some respect. For instance, the sample might have a

1 Comments

Survey-Weighted Tables and Graphs by using Stata

The June 2011 Granite State Poll included six questions related to global warming or climate change. Several questions were factual, but one (warmop) asked what you personally believe. Which of the following three statements do you personally believe? Climate change is happening now, caused mainly by human activities. Climate change is happening now, but

Bar Charts for Multiple Comparisons by using Stata

The catplot bar charts in Figure 4.3 depict a relationship between two categorical variables, each with four categories. If we have more than two variables, or more than a few categories, however, the catplot approach becomes cluttered. A cleaner alternative for making multiple comparisons of categorical variables employs Stata’s horizontal bar chart command hbar.

1 Comments