These pages look best viewed in Netscape!
Copyright 1998 by the Wyoming Department of Employment, Research & Planning


The Wyoming Wage Record Classification System

by: Brett Judd

The topic of how many people are working more than one job seems to interest a lot of people. A report (1), based on a survey of households, measured the number of multiple job holders in the state in 1996 at 9.5 percent. Multiple job holders are persons who work two or more jobs during a specified time period. This article will introduce a new way to count the number of multiple job holders. This new method, based on establishments (employers), measured the percentage of multiple job holders at relatively the same level of 9.9 percent. In addition to determining the number of multiple job holders, other workers in the state have been assigned a job classification.

Employers in the state submit their unemployment insurance (UI) reports to the Department of Employment on a quarterly basis. One of the reports is called wage records. Wage records contain a listing of each individual employee's social security number (SSN), his or her total gross wages for the quarter and the employer's UI account number. Wage records have been used in the past to track University of Wyoming graduates (2), to compare the wages of Wyoming's workers based on the industry they are working in and their gender(3) and to show how tenure can affect gender wages (4). Wage records can also be used for post-program analysis for such training programs as the Job Training Partnership Act (JTPA) or Vocational Rehabilitation. This article will present a new use for wage records and demonstrate how they can help obtain a better profile of the workers in the state.

After the wage records are collected from the individual employers each quarter, all of the different employers are combined together to produce one statistical file. The total number of records can vary depending on the quarter of the year. For instance, in 1996, the number of records (the total number of jobs worked) ranged from a low (222,016) in first quarter to a high (268,522) in third quarter (see Table 1). The records are then grouped together by the SSN with each employer and the corresponding wages listed in descending order. This means that in third quarter there were 268,522 jobs worked by 229,814 people.

Obviously, some people are working more than one job in a quarter. How many or what percentage of the people are multiple job holders? The answer is not as easy to determine as it might seem. It would seem logical to count how many people have wages from more than one employer and those would be the number of multiple job holders. When this is done, then there are 32,861 people working a second job in third quarter 1996 or 14.3 percent of the total workforce whose jobs are covered by UI stipulations. However, are they actually working both jobs at the same time? Or did they quit one job and then start a new job? It is not clear which of the two scenarios is true. The third quarter includes the months of July, August, and September. A person could work for one employer in July, not work for anyone in August, and work for another employer in September. In other words, the time frame within the quarter during which a person works cannot be determined from the data, only that they are working at some time in the quarter.

What is the percentage of multiple job holders in the state? To try to answer this question, a classification system for the wage record file has been developed. Instead of just concentrating on one quarter or each quarter individually, it was decided to look at the year in its entirety. The four quarter files for 1996 were merged into one file with one record for each SSN. When this was done, there were 283,101 unique SSN's (people who worked during the year). Initially, trying to determine the number of multiple job holders was the focus, but then it was decided that an attempt should be made to classify everyone. This was not an easy task. The classification criteria were revised many times and perhaps in the future they may be changed again, depending upon the usefulness of the current classification system.

There are six different groups or categories in the classification system. The organizing concepts are the degree of attachment to an employer and the labor market. The categories are: steady workers/same employer (those working at least three quarters for the same employer), steady workers/different employer (those working at least three quarters but not for the same employer), multiple job holders, job changers, two-quarter workers and one-quarter workers. In order to make the occurrence of a record mutually exclusive, certain requirements were established and inclusion in one group was considered in sequence before another. Since this project originated in order to determine the number of multiple job holders, priority was given to this category first. If a record met one of the conditions for a multiple job holder, then it was a multiple job holder first and would not be considered for any of the other categories. The next consideration was for the category of job changer, then a steady worker/same employer, then a steady worker/different employer. If a record did not fall into one of the above categories, then it would be either a one- or two-quarter worker. When necessary, the fourth quarter of 1995 and/or the first quarter of 1997 data were included to make a determination of the classification.

The classification system assigned each individual to a unique category. The totals for the categories are shown in Table 2, as well as the percent of the total. Most of Wyoming's workers fall in the steady worker/same employer category (127,982). The next largest category represents individuals who are only working one (45,210) or two (40,487) quarters during the year. Multiple job holders (40,186) comprise the next largest group and then job changers (18,512). The smallest group is steady workers/different employer (10,724).

The other categories may be analyzed at a later time, but for this article the focus is primarily on multiple job holders. Again, if the number of multiple job holders is divided by the total number of people, then the percentage is misleading. This gives the annual percentage of multiple job holders at 14.2 percent, or basically the same percentage as was listed earlier for third quarter. However, when during the year did these 40,186 people actually engage in multiple jobholding behavior? According to the definitions and classifications, these people are considered multiple job holders at some point during the year, but not necessarily the whole year. The number of multiple job holders that had at least two jobs during all four quarters is 5,333 (1.9% of the total), but this is only 13.3 percent of those who are in the multiple job holder group.

A more accurate count of multiple job holders for the year can be derived by determining the number of those people who are classified as multiple job holders and are working two jobs in a given quarter, then dividing that number by the total number of people working in that quarter. This will give the percentage of multiple job holders for each quarter. These figures are listed in Table 3. The percentage of multiple job holders varies each quarter with a high of 10.5 percent in third quarter to a low of 8.8 percent in first quarter. Then to obtain the percentage for the year, divide the average number of multiple job holders in each quarter by the average number of all workers each quarter. When this is done, the percentage of multiple job holders is reduced to 9.9 percent for 1996, which corresponds to the 9.5 percent found in the household survey.

This article focused primarily on multiple job holders while only introducing the other job classifications. Later this spring, Research & Planning will produce a separate publication about these classifications. The publication will provide the criteria that were used for selection into the different groups. It will also contain the demographic information for those people on the wage record file as well as employer information so that an analysis can be done with gender, age, wages and industry (primary or secondary jobs). The data for 1996 can also be compared to a file for 1993 to see if any changes or patterns exist. This is just a partial listing of the possibilities. This new classification system will be a useful tool--not only to measure multiple job holders--but also to look at other types of workers.


1 Please refer to "Multiple Jobholding" in the July 1997 issue of Wyoming Labor Force Trends.

2 Please refer to Tracking University of Wyoming Graduates Into the Wyoming Work-force, a report prepared for the Research & Planning Section of the Employment Resources Division, State of Wyoming.

3 Please refer to "The Relation of Age and Gender to Employment in Wyoming: Parts One and Two" in the May 1996 and June 1996 issues of Trends.

4 Please refer to "Gender, Tenure and Wages" in the August 1997 issue of Trends.

Brett Judd is an Economist, specializing in Labor Market Information (LMI) with Research & Planning.


Button Bar 4
Table of Contents | Labor Market Information | Employment Resources | Send Us Mail |



These pages designed by Gayle C. Edlin.
Last modified on October 1, 2001 by Valerie A. Davis.