PanelWhiz

From Wikipedia, the free encyclopedia

PanelWhiz is a data extraction tool which simplifies selecting and managing analysis data from large and complex longitudinal datasets. Panelwhiz is in use by economics and sociology researchers around the world.

PanelWhiz allows researchers to extract data from large-scale panel data sets using a graphic interface in Stata. Although the current package was created in 2006, it stems from previous software package called "SOEP Menu" which dealt exclusively with the complexity of the many data files that make up each wave of the German Socio-Economic Panel. The current package was developed by Canadian economist Dr. John P. Haisken-DeNew at the RWI Essen in Essen, Germany. Markus Hahn, an economics student at the Ruhr-Universität Bochum contributed to many components of the package. PanelWhiz is charityware, requiring the user to make a direct donation to UNICEF.

[edit] Basic Concepts in PanelWhiz

PanelWhiz uses the following definitions to create a structure under which PanelWhiz functionality can be described.

• a variable - typically all the respondents answers to a single question in a questionnaire.
• an item - a vector of variables identifying a single concept, such as "wage", but referring typically to many different waves.
• a project - a file containing a collection of items. "Wages", "employment" and "education" could be stored in a single project.
• a library - a directory containing a collection of projects. A library can be local or on the web.
• a plugin - a user-shared program that "cleans" an item for consistency over time
• a retrieval - a subset of data extracted from the source data sets, typically combining many years and levels of information from many different files.

PanelWhiz projects do not contain data, but rather only metadata, i.e variables and files names, but not the contents of the variables. This allows researchers to share Panelwhiz projects without breaking the data confidentiality constraints (no dissemination of unit record data) which are typically a legal requirement for users of panel survey data.

[edit] Plugins

PanelWhiz Plugins are an important component of the PanelWhiz package. In most longitudinal surveys, questions are improved and response possibilities change over time. In one year allowed responses may be (1) "yes" and (2) "no", but in following years, the response possibilities may be (1) "yes", (2) "maybe" and (3) "no". PanelWhiz uses its plugins to create response consistency over time. Plugins are also used to deflate nominal money values into real values. Plugins are contributed by PanelWhiz users and available to all other users.

[edit] Using PanelWhiz

PanelWhiz provides a webpage-like interface inside of Stata, allowing users to click not just on single variables but on vectors of variables, i.e. an item. This avoids having to lookup a variety of variable names stored in different files and different waves before extracting the data of interest.

The user starts Stata (once installed this also automatically starts PanelWhiz) and selects a survey. The user can open a particular wave of data (a single year) or select a keyword, such as "wages", allowing the user to search for items having the keyword in its label. The user finds the item of interest and clicks on the item. The variables for each and every wave pertaining to the item are then selected into the project. The user saves the project using a filename that he/she desires. The user can append other ready made projects to increase the number of items selected.

Given the information saved in the project, a data retrieval can be made, selecting only the variables found in the items in the project. A Stata do-file is generated and executed. A rectangularized file is created, merging all variables from any/all files. The data can be exported to other formats, such as SPSS, SAS, LIMDEP, GAUSS, and Microsoft Excel.

[edit] Supported data sets

• SOEP: The German Socio-Economic Panel
• HILDA: Household, Income and Labour Dynamics in Australia Survey
• CPS-NBER: Current Population Survey MORG from the NBER-Cambridge, Massachusetts
• CPS-CEPR: Current Population Survey MORG from the CEPR-Washington, DC
• IAB-BP: IAB-Nuremburg Betriebspanel (Firm Panel)

[edit] See also

• Stata
• Panel study

[edit] External links

• PanelWhiz - Homepage
• PanelWhiz User Manual - How to use PanelWhiz
• Panel Data - Known Panel and Repeated Cross-Section Data Providers
• RWI Essen - Rheinisch-Westfälisches Institut für Wirtschaftsforschung (RWI Essen), in Essen, Germany
• Extensive Introduction to Stata - Kit Baum (Boston College)

[edit] Research utilizing Panelwhiz

• Bauer, Thomas, Silja Göhlmann, Mathias Sinning (2006) "Gender Differences in Smoking Behavior" University of York, HEDG Working Paper 06/07 August 2006
• Bauer, Thomas K. and Mathias Sinning (2005) "The Savings Behavior of Temporary and Permanent Migrants in Germany", IZA DP No. 1632, Bonn
• Brenner, Jan (2006) "Parental impact on attitude formation: A siblings study on worries about immigration", MPRA Paper No. 34, Munich
• Corak, Miles , Michael Fertig and Marcus Tamm (2005) "A Portrait of Child Poverty in Germany", UNICEF / Innocenti Working Paper 2005-03
• Geishecker, Ingo, Holger Görg (2006) "Winners and losers: A micro-level analysis of international outsourcing and wages", DIW Working Paper August 2006
• Neumann Uwe (2006) "Intra-urban polarisation due to tertiarisation of the urban labour market? Evidence from urban regions in Germany," SCORUS 2006
• Tamm, Marcus (2007) "Does Money Buy Higher Schooling? Evidence from Secondary School Track Choice in Germany", ECINEQ 2007-58