Families on TANF in South Carolina: Employment Assets and Liabilities. Relationship Between Employment Liabilities and Employment Situation

10/01/2004

  • Personal barriers were more important than situational barriers in terms of current employment status.

The personal liabilities of respondents who were working at the time of interview looked quite different from respondents who were not working, as shown in Table V-1. Mental health and physical health problems, family members with health problems, possible learning disabilities, and lack of a high school diploma or GED were all more common among respondents not working at the time of interview than among those who were working.36

The logistical barriers we studied were found at an approximately similar incidence among those working and not working, except that more respondents who were not working reported transportation problems than those who were working. In this case, transportation could be an important reason those respondents were having difficulty becoming employed.

Table V-1
Percentage of Respondents with Employment Liabilities, by Current Employme nt Status (n = 1,120)
 
Employment Liabilities
Currently Working
Not Currently Working
Personal Liabilities Percent Percent
Physical health problem 11 27 **
Mental health problem 22 37 **
Child/family member with health problem 7 17 **
Severe physical domestic violence in past year
17 13
Possible learning disability 3 16 **
No high school diploma or GED 30 42 **
Difficulty with English 2 1
Criminal conviction 9 11
Logistical or Situational Liabilities Percent Percent
Child care problem in past year 26 27
Transportation problem in past year 27 33 *
Unstable housing in past year 22 22
Perceived problem in neighborhood 46 50
 
Source: Telephone Surveys of 1,120 TANF recipients in South Carolina
* Significant difference between employed and non-employed respondents at the 95% confidence level.
** Significant difference between employed and non-employed respondents at the 99% confidence level.
 
No relationship was found between current employment status and the other “personal” barriers including severe physical domestic violence in the past year, problems with English, and prior criminal record. In terms of logistical barriers, no relationship was found between current employment status and child care problems, housing instability, and neighborhood problems.
 
  • Personal barriers were also more important than situational barriers in terms of recent work history.
To understand the relationship between employment barriers and the longer-term employment situation of the respondents, we compared respondents who were working or who had worked in the past year to respondents who had not worked in the past year. We found that personal barriers were more significant than logistical barriers for both short-term and longer-term employability. Thus, ameliorating personal barriers may require longer-term and more complex solutions than addressing logistical barriers.
 
The specific findings are presented in Table V-2. In terms of personal barriers, there was a strong relationship between recent work history and physical health problems, mental health problems, learning disabilities, and lack of education.
 
Table V-2
Percentage of Respondents with Employment Liabilities, by Recent Work History (n = 1,120)
 
Employment Liabilities
Currently Employed or Worked in Last Year
Did Not mWork in Last Year
Personal Liabilities Percent Percent
Physical health problem 13 36 **
Mental health problem 27 41 **
Child/family member with health problem 29 39 **
Severe physical domestic violence in past year
17 11 **
Possible learning disability 6 21 **
No high school diploma or GED 35 43 **
Difficulty with English 1 2
Criminal conviction 11 9
Logistical or Situational Liabilities Percent Percent
Child care problem in past year 30 23 *
Transportation problem in past year 32 30
Unstable housing in past year 26 16
Perceived problem in neighborhood 48 50
 
* Significant difference at the 95% confidence level.
** Significant difference at the 99% confidence level.
Source: Telephone Surveys of 1,120 TANF recipients in South Carolina
 
There was also a relationship between recent work history and domestic violence but the relationship was not in the expected direction. Specifically, 17% of the respondents who had worked in the past year had experienced severe physical domestic violence compared to 11% of the respondents who had not worked in the past year.
 
The relationship between child care problems and recent work history was not in the expected direction in that 30% of the respondents who had worked in the past year reported that they had child care problems compared to only 23% of the respondents who had not worked in the past year. Some of the respondents who had not worked in the past year may not have been looking for work and therefore did not have problems arranging or paying for child care (or for transportation, for that matter).
 
  • The importance of personal barriers in current employment status was also apparent when controlling for other barriers and demographics.

Multiple regression analyses were conducted to examine the statistical significance of each employment barrier while controlling for the effects of other barriers and for the demographic characteristics of the respondents.37 The demographic variables in the analysis included age, ethnicity, and marital status.38

Table V-3 shows the results of the analysis of the relationship between the different independent variables and the current employment status of the respondents. As indicated, the variables that had the most negative impact on current employment status included (in order of importance, at the 95% confidence level):

  • signs of a possible learning disability;
  • child or other family member with a health problem or special need;
  • physical health problem;
  • mental health problem;
  • did not complete high school or GED; and
  • performed fewer than four common job skills.
As shown in Table V-3, severe physical domestic violence in the past year was significantly related to current employment status but not in the expected direction. That is, persons who had experienced domestic violence were more likely to be employed at the time of the survey than other respondents. The presence of a criminal record did not show a significant relationship with current employment status.
 
None of the logistical or situational barriers showed a significant relationship with current employment status (including transportation barriers, child care barriers, unstable housing, or neighborhood problems). In addition, none of the three demographic variables – age, ethnicity, or marital status – showed a significant relationship with current employment status.
 
Table V-3
Effects of Specific Variables on the Probability that a TANF Case Head Was Employed at the Time of the Survey
 
Independent Variable Coefficient
Significance Value
Demographics
30 years old or older -0.012 0.943
Black -0.024 0.886
Never married 0.098 0.553
Human Capital Liabilities
Did not complete high school or GED -0.355 0.016 *
Performed fewer than four common job tasks
-0.429 0.015 *
Personal Challenges
Physical health problem -0.604 0.004 **
Mental health problem -0.462 0.005 **
Child/other family member with health problem/need
-0.753 0.002 **
Physical domestic violence in the past year 0.428 0.013 *
Signs of learning disability -1.247 0.000 **
Criminal record -0.032 0.888
Logistical and Situational Challenges
Transportation barrier -0.213 0.174
Child care barrier -0.022 0.897
Unstable housing -0.060 0.721
One or more neighborhood problems -0.066 0.631
 
* Coefficient is statistically significant at the 95% confidence interval.
** Coefficient is statistically significant at the 99% confidence interval.
Source: Telephone Surveys of 1,120 TANF recipients in South Carolina

36 Differences were statistically significant at the 99% confidence level.

37 Variables such as limited English and chemical dependency were not included because very few respondents
reported any problems in these areas.
38 Binary logistic regression was used for the regression analysis. Binary logistic regression is similar to a linear
regression model but is suited to models where the dependent (outcome) variable is dichotomous. The predictor
variables (independent variables) can be either categorical or interval scale. If the predictor is categorical, a set
of dummy variables is created for use in the analysis. The predictor variables are analyzed and assigned a
coefficient that can be used to estimate odds ratios for each predictor in the model.

 

View full report

Preview
Download

"emp-assets-liab-223E.pdf" (pdf, 977.84Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®