Getting Started with STATA

This session will cover

• The software
• How to get data in
• Cleaning data
• Calculating descriptive statistics
• Making graph

The Software

Installing STATA

The interface

.do files

Installing STATA

• Stata is on all university IT managed computers on campus.
• University has a license to install on your personal laptop if you are in one of the departments in the Social Science Faculty
• To install, go to the IT software Stata pages

STATA Interface

• Open up STATA
• Results window (centre)
• Review window (left)
• Menus (top)
• Command window (bottom)
• The command is repeated as well as the answer in the results window
• Separate windows will appear for viewing data, plotting graphs, and writing commands

.do Files

• Type your commands and comments in a .do file.
• They are a record of what you have done. (There would be no record if you use the menus instead of commands.)
• To open a .do file either:
• Click pencil and pad icon for a new file or
• File > Open and navigate to a previously saved .do file
• Open the google doc and copy and paste the commands into the .do file
• Remember to save your .do file - this is not done automatically

Getting started

Using commands

Importing and viewing data

Basic commands

Running commands

• To run a command hold down ctrl and d:
• display “hello”
• display 4 + 4
• To add a comment use * or //
• Note the use of /* and */ for a multiple line comment.
• Use help command if you want more guidance on a function

Command notation

• Variable names are case sensitive
• # can be used to create a comment
• = is assigned as
• == is equal to
• != is not equal to
• < less than
• <= less than or equal to
• > greater than
• >= greater than or equal to
• & and
• | or
• ! not
• If can be used to specify certain conditions

Importing data

• You need to tell Stata what drive and folder the data file is in
• Use the command cd to change the directory (i.e. the drive and folder)
• cd “___DIRECTORY PATH___”
• Read in the data
• import excel using Name_Of_File.xlsx, firstrow
• Firstrow specifies that the first line is the variable names
• You can also use the menu options - File > Import > Excel
• You can read in other types of data (.csv, .sav, etc)

Viewing the data

• Variables window (top right)
• Properties window (bottom right)
• Click through variables to have a check
• To see data click grid with magnifier (top)
• To edit data click grid with pencil (top) (not recommended)

Some basic commands

• View your data:
• browse
• Check the dataset using describe
• describe
• Or use describe to check one variable in particular
• describe gdp
• To display the values of variables
• list

Cleaning data

Naming and labelling variables

Replacing values

Creating new variables

Naming and labelling

• To change the name of the variable
• rename debtgdpratio ratio
• STATA defaults to using the variable name as the label
• To change a label
• label variable nameofvariabletochange “Text to put in as label”
• Run the commands in the do file to change the labels
• Note: variable can be shortened to var

Replace a data point

• In STATA a missing data point is denoted by a .
• Browse data to check this
• The gdp for the UK is missing
• If you wanted to change a piece of data
• replace gdp = 2631.23 if country ==“United Kingdom”

Create a new variable

• You might want to use the data in your current variables to create a new variable
• For example you could create a variable of non European and European countries. 0 for non european and 1 for Europe
• Create a new column, with a name, filled in with 0s.
• generate european = 0
• Fill in with 1s if there is the word Europe in the variable region
• replace european = 1 if region == “Europe”

Number code a variable: Method 1

• Convert string variables into numbers - this makes it easier for the software to identify groupings and reduces the risk of human error
• For example 1 for Europe, 2 for Asia etc,
• gen regioncode= 0
• replace regioncode=1 if region==“Europe”
• replace regioncode=2 if region== “Asia”
• replace regioncode=3 if region=="North America"
• replace regioncode=4 if region=="Oceania"
• replace regioncode=5 if region=="South America"
• Add to the value label to note what these codes mean
• label define myvaluelabel 1 “Europe” 2 “Asia” 3 “North America” 4 “Oceania” 5 “South America”
• label values regioncode myvaluelabel

Number code a variable: Method 2

• Similar number coding can be achieved with
• encode region, gen (regioncode2)
• This variable has the data type long which is a number. STATA will treat this variable as a number.
• Note that the groupings are ordered automatically alphabetically

Calculate a new variable

• To calculate a new variable based on the values in current variables use generate or gen
• For example if you wanted to calculate the percent of debt for each country
• First calculate the total debt
• total debt
• Then you can use that total debt to calculate the percent of debt for each country in a new variable.
• gen debtpercent = (debt/45498.86)*100

Calculate a variable from multiple variables

• You can use information from several variables to create a new variable.
• You can also use criteria to select information
• For example use region and gdp to code countries as 1 for european gdp above $1000million, 2 for gdp below $1000million or 3 other countries.
• gen gdpeurope = 3
• replace gdpeurope=1 if region == “Europe” & gdp> 1000
• replace gdpeurope=2 if region == “Europe” & gdp< 1000

Selecting variables

• If you want to exclude a variable
• drop regioncode2
• If you want to exclude some rows/observations
• drop in 1 / 4
• Note drop is permanent!
• If you want to only work with certain variables
• keep country gdp ratio debt region income
• This will delete the other variables, and is permanent!

Save your data as a .dta file

• .dta files are STATA’s own data file format
• Save the data as a dta file e.g.
• save M:/Stata Workshop/gdpdebt.dta
• Next time you can read it in as a dta file using something like
• use M:/Stata Workshop/gdpdebt.dta

Calculating descriptive statistics

Summarising data

Temporarily selecting data

Creating frequency tables

Run summary descriptive statistics

• To calculate the number of observations, mean, standard deviation, minimum and maximum values use summarize (with a z) or sum
• For all variables
• sum
• You can just summarise one variable
• sum gdp
• You can get descriptive statistics according to criteria or groups
• sum if region == “Asia”
• sum gdp if gdp < 1000
• bysort region : sum gdp

Temporarily selecting data

• The preserve command remembers the dataset while you select a portion of it to analyse. You can then recall the preserved data.
• preserve
• keep if region == “Europe”
• summarize gdp
• restore, preserve

Creating a frequency table

• Use table or tab to make a frequency table - also known as a contingency table or crosstabs
• You can run these commands with one or two string variables
• tab region
• tab region income
• You can’t run tab on string variables - it gives you a table with each value listed which is not useful
• tab region gdp

Making graphs

Types of graphs

Different bar graphs

Changing how the graph looks

Types of graphs

• If you wanted to check for normality, create a histogram
• hist ratio
• A boxplot of gdp
• graph box gdp
• A scatterplot
• scatter gdp debt

Different bar graphs

• A bar graph
• graph bar gdp, over (region)
• A horizontal bargraph
• graph hbar gdp debt
• A horizontal stacked bargraph by region or country
• graph hbar gdp debt, stack
• graph hbar gdp debt, over (region) stack
• graph hbar gdp debt, over (country) stack

Label axes

• Start with a bar graph
• graph bar gdp, over (region)
• Add some x and y axes labels
• graph bar gdp, over (region) b1title(Regions of Countries in the Dataset) ytitle(Mean GDP)

Change the colours

• To make all the bars the same colour
• graph bar gdp, over (region) bar(1, color (green))
• To make each bar a different colour
• graph bar gdp, asyvars over(region) bar(1, color (green)) bar(2, color(red)) bar(3, color(blue)) bar(4, color(orange)) bar(5, color(yellow))
• Or use a preset scheme such as:
• s1mono
• economists
• s2color
• graph bar gdp, asyvars over (region) scheme(economists)

Save your graph

• With the graph still open
• graph export “My Graph.png”

Summary

Today, we have looked at:

• The software
• How to get data in
• Cleaning data
• Calculating descriptive statistics
• Making graphs

Resources

• Academic Skills Community Workshops
• One-to-one appointments with a maths or statistics tutor
• Digital Skills Training (including Library skills)
• Maths Skills Centre Practical Guides
• Statology Stata Guides
• Stata Cheat Sheets

Maths Skills Centre

• One-to-one appointments available to all students during term-time and vacation
• Personalised advice and guidance on maths and statistics topics
• Online and on campus workshops covering a range of topics including statistical tests and software

Maths Skills Centre

Any questions?

Please book a one-to-one statistics appointment if you have questions you would like to discuss with a statistics tutor.

If you have any feedback you would like to share with us, please fill in this feedback form.