Posts

RSG Logos

New for Fall 2015! Half-day and full-day Advanced Software Training

, ,

New! CTRL’s Research Support Group is offering extended software training sessions. For Fall 2015, we are offering full-day training in Stata and R, and half-day training in Qualtrics and NVivo. The full-day training sessions will include a light lunch.

Full descriptions for the workshops are below. If you are interested in registering to attend these, or any of our introductory or intermediate level software trainings, please register at the CTRL website: www.american.edu/ctrl/rsgevents.cfm.


 

Use of Stata for Data Management, Analysis, and Visualization
Friday, October 2nd, 10:20 am – 3:50 pm (Light lunch will be provided)
This one-day workshop offers opportunity to users, who already have some familiarity with this statistical application, to learn about a variety of advanced topics and improve their grasp on utilizing the potential of the software. Selected content will include, but is not limited to, programming with Mata, simulations, structural breaks, time series and panel data, discrete choice models, and tips and tricks in coding. Following a brief refresher on introductory aspects of Stata, the workshop will consist of about one hour allocations to discuss the topics listed above. Emphasis will be given to the coding component of working with the program and not so much to the underlying theoretical econometric principles. Register here.

Use of Qualtrics and SPSS for Survey Research and Data Analysis
Thursday, October 8th, 1:10 pm -5:15 pm
Qualtrics is a professional-level online survey software available to all AU faculty, staff, and students, and is frequently used by faculty in both research and teaching contexts. Attendees will learn the basics of online survey creation, as well as advanced methods for organizing and distributing surveys. After working on survey creation, we will explore the data analysis functions internal to Qualtrics, and show how to export survey results to SPSS and perform more sophisticated multivariate data analyses and visualization. Register here.

Use of NVivo for Qualitative Research
Thursday, October 15th, 1:10 pm – 5:15 pm
NVivo is a software tool for undertaking qualitative research that is available to all members of the AU community. We can install it on personal computers and it is also available through the Virtual Computer Lab. We plan on showing some of the basic features of NVivo in importing materials and analysis of them, but want this training to be a working session where participants bring their own qualitative data (text, audio, or video) and actually work on a project. Because there are so many different types of qualitative approaches, we want those attending to come with their own methods in mind and data particular to exploring a relevant research topic. We invite those who are thinking of attending to come by the CTRL Research Lab and get the software installed on their computer prior to the session. We can also advise on proper formatting of documents so they are easy to import into NVivo. Register here.

Use of R for Data Management, Analysis, and Visualization
Friday, November 13th, 10:20 am – 3:50 pm (Light lunch will be provided)
This is a half-day course that guides participants through the fundamentals of using R software for a typical data analysis process. Participants will learn the basics of reading data, descriptive statistics, data visualization, data analysis, and presentation of analytical results. In addition to the fundamentals, faculty may opt to learn some advanced use R techniques, such as time series, panel data, survival regression analysis, as well as utilization of R for text mining, parallel computing, visualization and analysis. Register here.

Fall in love with SPSS again: Mapping in SPSS 20

, ,

SPSS manages to keep their throne as the most widely-used statistical analysis software in social science. They have outdone themselves, yet again, with SPSS 20.

The Maps feature is at the forefront of the collection of new updates in this recent release. Now, you can incorporate geography into your quantitative research, in order to visualize geographical observations or trends.

This feature is contained in Graph > Graphboard Template Chooser. There are three types of map templates:

  1. Choropleth Maps, where you can use color gradation to reflect the value of statistic on a variable (mean, median, and sum).
  2. Mini Charts, displaying a chart for certain regions; each labeled with the respective location.
  3. Overlay Maps, combining two map files into one visualization; one as reference, another for data display.

You can work with the following data:

  1. Latitude and Longitude, in order to generate a reference map with geographical coordinates.
  2. Existing map files: ESRI shapefiles easily converted with the Map Conversion Utility
  3. Pre-installed maps in SMZ format, including the U.S. map with states and cities indicated

For additional map files: “Many of the templates in this product are based on publicly available data obtained from GeoCommons (http://www.geocommons.com) and the U.S. Census Bureau (http://www.census.gov). Another source for U.S. federal, state, and local geospatial data is the U.S. Geological Survey (http://www.geodata.gov).”

There are various ways to customize your map to display the data as you desire, including color schemes (stylesheet) as well as Pie on a Map.

Pie on a Map

Reference Map with Latitude and Longitude

SPSS 20 is now available on all the computers in Hurst 202 and 203.

References:

IBM. SPSS. Program documentation. Converting and Distributing Map Shapefiles. Vers. 20. IBM, 2011. Web. 14 Feb. 2012. <http://publib.boulder.ibm.com/infocenter/spssstat/v20r0m0/index.jsp?topic=%2Fcom.ibm.spss.statistics.help%2Fmapconversion_intro.htm&gt;.

PSPP: SPSS’s Alter Ego

, ,

You’re considering getting your own copy of SPSS on your personal computer and you begin to browse for the program online (although if you have an AU-owned computer, you can just get it installed from us).  You want more than the free trial version that the makers of SPSS offers but you don’t want to pay the price. What to do? There is an option (a FREE option, if we may add) out there that may help: PSPP.

What is PSPP?

PSPP is a free software application for statistical analysis. It is intended to be a free replacement of the proprietary program SPSS. When you open PSPP, you’ll find that the interface looks very similar to SPSS with a few exceptions. All functionality that PSPP currently supports is in the core package.

PSPP has a data view tab (spreadsheet), a variable view tab (to create variables and define their characteristics) and has an easy to use point-and-click interface.

From the PSPP website: PSPP can perform descriptive statistics, T-tests, linear regression and non-parametric tests. Its backend is designed to perform its analyses as fast as possible, regardless of the size of the input data. You can use PSPP with its graphical interface or the more traditional syntax commands.

Again, PSPP is very similar to SPSS.  Even the output window is comparable to SPSS.

A brief list of some of the features of PSPP follows:

  • Supports over 1 billion cases.
  • Supports over 1 billion variables.
  • Syntax and data files are compatible with SPSS.
  • Choice of terminal or graphical user interface.
  • Choice of text, postscript or html output formats.
  • Inter-operates with Gnumeric, OpenOffice.org  and other free software.
  • Easy data import from spreadsheets, text files and database sources.
  • Fast statistical procedures, even on very large data sets.
  • No license fees.
  • No expiration period.
  • No unethical “end user license agreements”.
  • Fully indexed user manual.
  • Free Software; licensed under GPLv3 or later.
  • Cross platform; Runs on many different computers and many different operating systems.

Visit PSPP’s website to learn more about the software and information on how install it.  Don’t forget to look closely at the different installation instructions for PCs and Macs.

SPSS Tutorial Videos on YouTube

, , ,

Need some last-last-minute help with SPSS?

California State University, Los Angeles, has a great YouTube channel that has several videos on how to use SPSS.  There are videos on how to download data files, define variables, entering data, and how to do various analyses, like crosstabs, t-tests, regressions, and more.

Below is an example of one of their videos. This one is a basic intro to SPSS:

 

 

This channel is also really handy if you need a refresher on how to use SPSS.  Most of the videos are between 4 and 5 minutes long, so they’re also really good if you need to remember how to do that one thing that slipped your mind. We have a more recent version of SPSS in the lab, but the steps to do these analyses are the same! Ask a consultant for help if you need additional assistance.

Quick Way to Find Variables on SPSS

When you’re using SPSS, have you had those moments when you’re endlessly looking for a certain variable, but in a list of 100+ you never seem to do it quickly?  Do not waste time scrolling through your giant database; instead, take this tip from us:

First, remember that SPSS stores your variables with a name (= a code for the variable, which should only be a word) and a label (= an explanation of what your variable is in a whole sentence). Second, whenever you want to do an operation in SPSS (i.e. get descriptive statistics for a variable) the window for this will show a list of the variables that by default presents the labels.  If you point the cursor over this list (anywhere on this list of variables is fine) you can right click the list and choose that you want to see the “names” of these variables.  If you right click again you can indicate that you want to see an “alphabetical sort” of the variables.

 And done!  We suggest looking at variable names because is just a word and is likely that you would know this, plus the alphabetical sorting tends to make more sense with the variable name rather than the code.

Please leave a comment if you think this was useful, and ask a consultant if you have a question using SPSS!

SPSS vs. STATA

,

SPSS vs STATA

Are you still trying to decide which statistical package to use for your research project? Take a quick tip from one of our consultants to help make your decision:

These are some of the pros and cons of using SPSS or Stata for your research analyses.

SPSS

SPSS is good for managing basic datasets and if you are not going to do cutting-edge statistical analysis.  Its biggest advantage is that it is user friendly because its layout is so much like the familiar Excel spreadsheet.

Although both SPSS and Stata have a menu-driven option, the point and click menus in SPSS are the easiest to learn. You could say that SPSS has moved away from an academic research focus and seems to invest more of its development in marketing-oriented graphics. Many NGO’s, banks and related type entities would most likely not use SPSS for their work and would choose Stata over it.

SPSS’s graphics are very professional and they look appealing. The great thing about graphics generated on SPSS is that you can edit them as you please (colors, labels, size, etc).

SPSS screenshot

Stata

This program is widely used in NGO’s, banks and many other organizations that require data analysis. It is best for complex statistical analyses. Unlike SPSS’s “drop-down menu” feature, Stata tends to be more “command-based”.

The first window you encounter will not be the friendly table layout that you may be used to seeing in Excel. The layout may catch you by surprise, but once you get used to what each window tells you, it becomes fairly easy to use. It has a command window in which you can type exactly what you want to do (i.e. for a regression between two variables named “infmort” and “doctors” you would type: “reg infmort doctors”, hit Enter and your regression analysis pops up).  It is very intuitive and produces very professional output tables.

Stata screenshot

If you can’t decide which program to use for your analysis, consider even using both.  As always, ask a consultant for help if you need further assistance!

A Beautiful Union: How to merge data on SPSS

You’ve just spent hours finding the perfect data for your project. The problem now is that you have two (or more) different data sets and you need to smoosh them all into one. Here is how you can do that:

 

YouTube video: California State University, ITS Training Program

1. First, the two data sets you want to merge need to have a common, unique identifier (key variable) for each case in your data set. Both data files should provide different data for the same set of cases. Also, make sure the cases are spelled exactly the same and are in the same order in both data sets (You could try sorting them in ascending order before trying to merge).

2. Open the first file that you wish to merge. Under the “Merge Files” item in the Data menu, select “Add Variables”

3. Select the file you wish to merge

4. If the cases have the same name one will show up in the “Excluded Variables” list. This will be the common variable that you will use to merge the data files together. If your key variables have different names you should rename one so they both have the same name. Select the key variable from the “Excluded Variables” list and check the “Match cases on key variables in sorted files” checkbox. Then send the variable to the key variables box (using the arrow pointing to the right).

5. Click OK. SPSS will give you a warning regarding sorted key variables. If everything is in order (make sure both files have key variables that are spelled the same and are sorted in order before trying to do a file merge) click OK again. All variables from both files will be merged in a new datafile, and cases are matched by the key variable.

As always, ask a consultant for help if you get lost!

Things to keep in mind while using SPSS

When using SPSS, you’ll quickly find that there are few conventions unique to this software, especially when it comes to naming variables and recognizing your files after you’ve saved them. Here are a few things to keep in mind when using SPSS:

File name extension:
Data files are given a “.sav” extension while output files are given a “.spv” extension. If you open an output file to launch SPSS, it will open an empty data file with it.  You will need your data to continue working on your output file. Make sure you save both your data and your output files!

Variable names:
Variable names are limited to 8 characters. SPSS won’t allow you to use characters such as !,@,#,&, etc. You can use an underscore (_) but no spaces. Also, make sure when you name your variable it is recognizable. It will save you from being confused later.

Importing from Excel:
A neat feature on SPSS is that you can import data from an Excel spreadsheet onto SPSS. When doing so, make sure you search for the file with the “.xls” extension.

Output files:
You can copy the output you’ve generated from your data by simply clicking the table or chart, copying, and pasting it in another application.  You can also print a portion of your analysis from the output file by highlighting the specific portion of the file that you want to print.

Sorting variables:
Don’t forget that when you are performing an analysis, you can sort your variables alphabetically by variable name or label name by right clicking on the variables. This will save you a lot of time when searching for the variables you want!