Immigrant Construction Workers and Safety and Health in South Florida

TABLE OF CONTENTS: Introduction. Methodology, Characteristics, Empirical Results Relationships of Variables with Safety Outcomes Summary, Conclusions, Appendices, References

A Research Report

Bruce Nissen
Research Institute on Social and Economic Policy (RISEP)
Center for Labor Research and Studies
Florida International University
June 2007

Acknowledgements

I would like to thank Carmen Figueredo for translating the survey instrument into Spanish and Yuset Cueto for re-translating it back into English, to ensure equivalence of research instruments. I also thank Alejandro Angee, Emily Eisenhauer, Marcos Feldman, Eduardo Granados, Cynthia Hernandez, and Lina Stepick for persistent and excellent surveying of south Florida immigrant construction workers. I thank Yue Zhang for inputting the data into a usable database and producing many of the tables, and Alex Angee for producing the charts.

Dale Belman kindly shared with me a construction worker survey he was developing. This survey helped me greatly in formulating questions for my own survey. I thank Jim Platner of the Center to Protect Workers’ Rights for sharing his vast knowledge of the literature and the subject with me as I undertook this project.

Finally, I acknowledge the Center to Protect Workers’ Rights for their financial assistance without which this study could not have been completed.

Contact information:

Bruce Nissen, Director
Research Institute for Social and Economic Policy (RISEP)
Center for Labor Research and Studies
Florida International University
University Park
Miami, FL 33199
Ph: 305-348-2616
Fax: 305-348-2241
E-mail: Bruce.Nissen@fiu.edu

IMMIGRANT CONSTRUCTION WORKERS AND SAFETY AND HEALTH IN SOUTH FLORIDA

EXECUTIVE SUMMARY

This report is based on surveys conducted with four hundred construction workers on large construction sites in South Florida in 2006. The survey elicited information on the safety training received, personal protective safety equipment regularly used, and the safety policies and practices of the employers of these workers. It also collected a wide variety of demographic data and information on non-safety employer practices that were thought to be possibly related to safety conditions on the job.

The purpose of the study is threefold:

1. To provide a portrait of south Florida immigrant construction workers: demographics, incomes, safety conditions on the job, and employer treatment in other ways that may be related to their safety conditions;
2. To determine if union membership or documented legal status of these workers is associated (perhaps causally) with different safety outcomes (measured by degree of safety training, use of personal protective equipment, and employer safety policies and practices) and secondarily to see if other factors may be associated with better or worse safety conditions; and
3. To analyze results to determine any public policy measures that may improve the safety conditions of these workers.

Key findings

DEMOGRAPHICS: Virtually all immigrant construction workers were from Central or South America or Mexico or the Caribbean. Cubans, Nicaraguans, Mexicans, and Hondurans were most numerous. They averaged thirty-six years old, and all but two were male. A majority had been in the U.S. ten years or less; the average (mean) number of years was twelve. They averaged approximataely seven and a half years of work experience in the U.S. construction industry. Almost a third earned less than $20,000 per year, and half had a family income below $30,000 per year. (Average family income for all Florida residents was approximately $55,000 per year). Forty-four percent had less than a high school degree; yet twenty-one percent had taken some college courses and thirteen percent had a college or graduate degree. Seventy-three percent were either a U.S. citizen or had a documented legal status, while twenty-seven percent were undocumented.

SAFETY TRAINING: Seventy percent had received the basic safety training for the industry, the “OSHA 10-hr. training;” an almost identical percentage had received training in scaffold safety. Other types of training (CPR/First Aid, Asbestos Awareness, Hazardous Materials) had been provided to between one fifth and fifty-seven percent, depending on type of training.

USE OF PERSONAL PROTECTIVE EQUIPMENT: Use of work boots, hard hats, and protective eyewear was nearly universal. Between forty-two percent and seventy-eight percent regularly use other types of protective equipment (work gloves, guards on cutting tools, hearing protection, respiratory protection), depending on type of equipment.

SAFETY POLICIES AND PRACTICES OF EMPLOYERS: Some employer practices, like required use of body harness, providing hand rails for scaffolds, and provision of drinking water and bathrooms, were virtually universal. Other practices, like holding weekly safety meetings, providing copies of a safety program, providing access to Material Safety Data Sheets (MSDS), providing ground fault electrical outlets, and providing first aid kits, were practiced by employers between half and ninety percent of the time, depending on particular practice. Over a fifth used cut and taped electrical cords at the worksite.

INJURIES: Eleven percent had experienced a workplace injury within the past three years that resulted in loss of work of a day or more. Thirty-nine percent had witnessed a worksite accident within the past year serious enough to cause a fellow worker to be taken to the hospital. In their entire construction work career (average length: approximately seven and a half years), eighteen percent had witnessed a death at a worksite where they worked.

OTHER EMPLOYMENT CHARACTERISTICS AND PRACTICES: A majority had worked for their current employer less than one year, but most had worked for only one or two employers in the past twelve months. Nine percent worked for a temporary help firm, not a construction contractor. Sixty-six percent worked for an employer with one hundred or more employees and more than a quarter worked for an employer of five hundred or more. Eighteen percent had been paid at some point in their construction career in cash, and eight percent had been asked to dishonestly sign a form (a “1099 form”) stating that they were an independent contractor rather than an employee. Thirty-five percent had been offered a retirement or savings plan, and almost fifty-six percent had been offered health insurance coverage.

PERCEPTIONS OF EMPLOYER ATTITUDES ABOUT SAFETY: Over ninety percent thought that their foremen and employers were concerned about safety and that their job site had a good safety program. Ninety-six percent were willing to report a safety violation that they saw; the other four percent were afraid to do so. Eighty-nine percent thought that unions lead to safer jobs. However, close to twenty percent thought that productivity was more important than worker safety at the place they work, and sixty percent thought that their work conditions were dangerous.

RELATIONSHIP OF UNION MEMBERSHIP AND DOCUMENTED LEGAL STATUS WITH SAFETY OUTCOMES: It was hypothesized that union membership and documented legal status are associated with better safety outcomes than those experienced by their non-union and undocumented counterparts. Results show the following:

• Initial cross-tabulations broadly confirm both hypotheses, but documented legal status is unexpectedly associated with less use of personal protective equipment.
• However, these results may be spurious if “union member” and “documented” are acting as proxies for other factors, such as length of time in the industry, length of time in the country, or industry craft (skill). After performing a variety of tests for the possibility of proxy behavior, results suggest that only unionization is consistently associated with improved safety outcomes. Documented legal status loses all its association with better safety outcomes when only one craft (for example, carpentry) is considered.
• The positive union association with improved safety outcomes is most pronounced for those immigrant construction workers with shorter time in the industry.
• Union membership is also associated with sharply lower serious injury rates than those of non-union immigrant workers. And, when union members are injured, they lose far fewer days of work than do their non-union counterparts. (Caution: these results come from a relatively small sample on the union side, making results only preliminary.)
• The associations found probably show a positive union “impact” given intuitive knowledge about the plausibility/possibility of causality in each direction.

OTHER ASSOCIATIONS DISCOVERED:

• Completely unskilled (general laborer) respondents were less likely to receive safety training than were their semi-skilled or skilled counterparts. There is some very weak evidence that they may experience inferior employer safety practices, but no evidence that they use less personal protective equipment on the job.
• Immigrant construction respondents who were either (1) paid in cash, (2) not provided a health insurance plan, or (3) not provided a pension plan received less safety training and experienced worse employer safety practices. Less favorable (no health plan, no pension) or irregular (cash payment) practices of a non-safety nature are “bundled” with inferior safety treatment. (“Bad” employers “cut corners” in all areas, including safety.)

POTENTIAL PUBLIC POLICY IMPLICATIONS:

• Unionization should be encouraged if the aim of public policy is to improve the safety conditions of these relatively vulnerable workers in a very dangerous industry.
• Public policies that encourage or require better treatment in areas like employer-provided healthcare and pension plans may improve the safety of these workers, either through the mechanism of “weeding out” the “bad” employers who skimp in all these areas (including safety) or by forcing employers to develop a more responsible attitude toward employee treatment in general. Similarly, perhaps stronger enforcement of wage and hour laws to ferret out illegal cash payments in the underground economy would lead to safer work for these immigrant construction workers. But these conclusions are very tentative because the research only uncovered a positive association between these undesirable non-safety employer treatments and worse safety outcomes; it did not find or prove a causal relationship between them and less safe conditions.

INTRODUCTION

Immigrants are a large and growing percentage of the total construction labor force in the United States today. According to the Current Population Survey (CPS) Annual Social and Economic Supplement, as of March 2006 almost twenty four percent of all construction workers in the country were foreign born. Most of the immigrant construction workers are Hispanic, although not all Hispanic construction workers are immigrants, of course. In March 2006 a little over twenty-four percent of the construction work force was Hispanic, compared to approximately six percent in 1980 (Construction Chart Book: Chart 16b). Seventy percent of the 1.4 million Hispanic construction workers in the U.S. in 2000 were born outside the United States, and fifty-seven percent were not U.S. citizens (Construction Chart Book: section 16).

Immigrants and Hispanics are an even larger percentage of the construction workforce in Florida than they are nationally. According to the CPS Annual Social and Economic Supplement of March 2006 immigrants were 34.7% of Florida construction workers, and Hispanics were 31.69%. And south Florida has an even greater concentration: about three quarters of construction workers in Miami-Dade County (Miami area) are Hispanic.

Hispanic workers (and most likely immigrant workers) in this industry face especially dangerous working conditions. Hispanics constituted less than 16% of the construction workforce in 2000 yet suffered 23.5% of fatal job injuries. Hispanic construction workers that year were nearly twice as likely to be killed by occupational injuries as their non-Hispanic counterparts (Dong and Platner: 2004).

For that reason it is important to investigate the conditions these workers face, both because they are an ever-growing segment of the workforce and because their treatment will affect the treatment of all U.S. construction workers. This study explores the safety and health training and the safety and health conditions of immigrant construction workers in south Florida. It has several purposes. First, it provides a general portrait of who these workers are, where they come from, length of time in the country, and the like. Second, it gives a general picture of the safety and related conditions of these workers, providing a preliminary picture of how they are trained and treated in the area of safety and health. Third, it looks for relationships between other statuses/conditions of these workers and their safety training and conditions. And finally, it offers tentative public policy measures that may improve the safety of these workers.

The following section examines literature relevant to the present study. Following that, the methodology of the current study is explained. Then a section summarizes the characteristics of those surveyed. The section following that displays the results from the survey answers, with a minimum of analysis or interpretation. Two sections after that present some hypotheses about likely factors influencing the different safety and health outcomes for different workers, followed by a testing for relationships that provide evidence for or against those hypotheses. Finally, a concluding section summarizes and discusses the results and offers public policy suggestions.

LITERATURE ON THE TOPIC

An earlier pilot study of immigrant construction workers about safety and health issues by the author surveying only fifty workers in south Florida was done in 2004 (Nissen, 2004). That study found that the surveyed workers were primarily Hispanic and that most were not U.S. citizens, although a majority was legally documented. It found that they labored under extremely unsafe conditions, had less than adequate training, generally used personal protective equipment but had less consistent employer safety policies and practices, and sometimes faced questionable or illegal employer practices making medical care for serous injury difficult. It found only two factors consistently associated with positive safety outcomes (measured by safety training, use of personal protective equipment, and safer employer policies and practices): union membership and documented legal status. Other potential correlations with better safety outcomes, such as longer residence in the U.S. or longer tenure in the U.S. construction industry, were not found to be significant. This study was suggestive but not definitive because of small sample size.

Some studies have done a comparative analysis of injuries or illnesses of Hispanics vs. other groupings, such as non-Hispanic whites and blacks. Robinson (1989) surveyed California data and discovered that for all workers (not specifically construction workers), Hispanic workers faced higher probabilities of exposure to occupational injuries and illnesses than did non-Hispanic whites. Utilizing emergency room records and looking at construction workers in the Washington D.C. area, Hunting, Nessel-Stephens, Sandford, Shesser, and Welch (1994) found that laborers and Hispanic workers were over represented among severe cases of injury. Looking at New Jersey construction workers, Sorock, O’Hagan Smith, and Goldoft (1993) found that Hispanics had death rates over three times that of non-Hispanic whites. Anderson, Hunting, and Welch (2000) found that Hispanic construction workers were more likely to be employed in the less-skilled trades and had a higher proportion of serious injuries. They suggested that minority status is a predictor of trade and that trade is a predictor of injury risk. Welch, Hunting, and Nessel-Stephens (1999) found that Hispanic and older construction workers were more likely to have continuing symptoms long after an injury. The Dong and Platner (2004) study cited in the introduction found that from 1992 to 2000, for every age group, Hispanic construction workers consistently faced higher relative risks. All of these studies suggest that Hispanics in the construction industry are more likely to face injury and inadequate safety conditions than others.

O’Connor, Loomis, Runyan, Abboud dal Santo, and Schulman (2005) surveyed fifty young Latino construction workers. This study was concerned with both their youth and their ethnic status, and concluded that they had received very little health and safety training, particularly those with less English language ability.

Of course, not all Hispanic workers are also immigrants. Very few studies have been done looking specifically at health and safety conditions of immigrant construction workers in the U.S. although there are some regarding immigrant workers in other or all occupations or in other countries (Gannagé 1999; Wu, Liou, Hsu, Chao, Liou, Ko, Yeh, and Chang 1997). Perhaps closest to the aim of the present study, Pransky, Moshenberg, Benjamin, Portillo, Thackrey, and Hill-Fotouhi (2002) surveyed urban immigrant workers in an immigrant community in northern Virginia, and found that they face high risk of occupational injuries, with adverse outcomes. Thirty-two percent of these workers worked in construction, and of that group, thirteen percent had been injured in the past three years.

A small number of studies have been done on the impact of unionization on workers’ safety. Taylor (1987) found that the degree of unionization in an industry (not only the construction industry) and its safety record was significantly positively correlated in some years but not in others. He explains these differences in terms of a number of intervening variables, including labor-management safety committees and safety consciousness of union members or management. He thus finds the relationship between unionization and safety to be complex. Dedobbeleer, Champagne, and German (1990) studied construction workers in the Baltimore area and found that union membership is significantly positively correlated with high safety performance. However, controlling for age (age 26 or younger vs. ages 27 and up) made most of the relationships insignificant, since union workers tended to be older. However, there was an extremely high correlation between union membership and exposure to safety training. This correlation remained significant after all attempts to control for all other variables. Yet, they found that the differences in likelihood of being injured were in the expected direction (union worker injury rates were lower), but not significant.

While these studies are suggestive of unique safety and health issues and problems for immigrant construction workers in the United States, none of them apart from the author’s pilot study directly attempt to discern factors that might influence the safety and health outcomes for this population. This study empirically attempts to discover the safety and health conditions of immigrant construction workers in south Florida to determine if there is a relationship between the different safety outcomes they experience and other factors captured in the survey, and to compare their safety outcomes to those of non-immigrant construction workers employed at south Florida worksites.

METHODOLOGY OF THE CURRENT STUDY

Four hundred construction workers in Miami-Dade County, Florida were surveyed in the summer of 2006 using a sixty question survey instrument constructed by the author. The total sample was composed of 283 immigrant workers and 117 non-immigrant workers employed alongside the immigrant sample. All construction workers in the sample pool were given identical surveys.

The survey instrument asks questions concerning demographic data, safety training, workplace safety practices, employer safety policies and practices, other employer practices regarding wages, pensions, workers compensation, and respondent evaluations of their employers’ attitudes toward safety. Workers were surveyed in Spanish or English, depending on the language preference of the person being surveyed. Surveyors were fluent in Spanish and English. The original English language version of the survey instrument was translated into 5 Spanish and then re-translated back into English by different individuals, to ensure equivalence of survey instruments. (Copies of the survey instrument in English and Spanish are attached to this report as Appendix A and Appendix B.)

It is impossible in a project of this nature to get an entirely random sample of the universe of south Florida immigrant construction workers. There is no database containing the names and contact information for such workers. The researcher approximated random selection processes as closely as possible by creating a database of all Miami-Dade County construction projects costing over $10 million derived from the Dodge Report (containing “open bidding” projects) and the Industrial Info Report (containing “closed bidding” projects). Thus, the universe which was sampled was comprised of medium- and large-sized construction projects in the county. Sites were randomly selected from this database, and surveyors were sent to them to contact workers before and after the workday.

At the sites, surveyors sampled workers either as they prepared to begin work or as they ended their workday. In addition, through a “snowball” technique, participating workers sometimes led surveyors to other workers willing to participate in the survey. Thus individual respondent selection at the site was as close as one can come to random selection. (To get a random sample of the workers at the site would require the contractors’ and sub-contractors’ cooperation, which would introduce an enormous employer “self-selection bias” between those willing to cooperate and those unwilling). No more than 20 workers from any one site were surveyed to ensure a large and representative set of sites; however, since the local construction workforce is less than 5% unionized, an “all union” construction site was over-sampled to ensure a large enough number of union workers to be able to make meaningful comparisons. The researcher aimed to include at least twenty percent union members in the sample.¹

Surveyors for this research project were Florida International University graduate research assistants and personnel with previous training in social science methodology. All surveyors were given additional training specific to the use of this particular survey instrument. Informed consent was obtained in accordance with the research protocols of Florida International University. All survey respondents were given a nominal sum of $20 as a token of appreciation for their cooperation.

Results were collected and entered into SPSS (statistical software) and the resulting database was analyzed to determine a number of demographic facts about these workers. As the following sections will show, an analysis was done of relationships between safety outcomes and a variety of other factors.

CHARACTERISTICS OF THOSE STUDIED

This section will cover only the characteristics of the immigrant workers surveyed. The native born were only included as a control group; this report is confined to results from the sample of immigrant construction workers.

Four countries comprise almost seventy four percent of the countries of origin for these workers. Cuba supplied over one fourth, while Nicaragua, Mexico, and Honduras each supplied around fifteen to eighteen percent. Virtually all were from Central or South America or the Caribbean. Table 1 shows details.

Table 1 Country of Origin of Immigrant Construction Workers Surveyed

Country	Number
Cuba	74
Nicaragua	50
Mexico	42
Honduras	42
Haiti	11
El Salvador	10
Guatemala	9
Colombia	8
Brazil	6
Peru	5
Puerto Rico*	5
Dominican Republic	4
Trinidad & Tobago	3
Venezuela	3
Bahamas	2
Canada	2
Jamaica	2
Ecuador	1
Grenada	1
Panama	1
St. Vincent	1
Total	282

One immigrant did not answer this question.
*For the purposes of this study, Puerto Ricans are counted as “immigrants” even though technically they are not, since Puerto Rico is a U.S. territory.

All but two of the 283 respondents were male, with the two females being a 34 year old woman and a 35 year old woman, both from El Salvador. Respondents averaged 36 years of age, ranging between a 17 year old Mexican and a 78 year old immigrant from the Dominican Republic. Table 2 shows the spread of ages, in increments of ten.

Table 2 Age of Immigrant Construction Workers Surveyed

Age	Number	Percent
17-19	12	4.3%
20-29	85	30.1%
30-39	84	29.8%
40-49	58	20.6%
50-59	35	12.4%
60-69	7	2.5%
70 and over	1	0.4%
Total	282*	100.0%

*One respondent did not answer this question.

On average, respondents had resided in the United States 12 years, with a range between less than a year (eleven people) to 63 years. Over fifty-six percent had been in the country ten years or less. Table 3 shows the spread, in increments of five years.

Table 3 Years of residence in the U.S. of Immigrant Construction Workers Surveyed

Years	Number	Percent
0-5	98	34.8%
6-10	60	21.3%
11-15	30	10.6%
16-20	37	13.1%
21-25	19	6.7%
26-30	22	7.8%
31-35	6	2.1%
36 and over	10	3.5%
Total	282	100.0%

One respondent did not answer this question.

They averaged 7.35 years working in U.S. construction, with a range from one week to 57 years. Most were concentrated at the lower end of the spectrum. Table 4 shows the spread, in increments of three years.

Table 4 Years of U.S. construction work of Immigrant Construction Workers Surveyed

Years	Number	Percent
0-3	121	42.9%
4-6	55	19.5%
7-9	30	10.6%
10-12	22	7.8%
13-15	15	5.3%
16-18	13	4.6%
19-21	11	3.9%
22-24	1	0.4%
25-27	7	2.5%
28-30	1	0.4%
30+	6	2.1%
Total	282	100.0%

One respondent did not answer this question.

The primary trade of these workers was carpenter, followed by ironworker and general laborer. Table 5 shows the results for all trades represented.

Table 5 Primary Trade of Immigrant Construction Workers Surveyed

Primary Trade	Number	Percent
Carpenter	100	35.3%
Iron Worker	46	16.3%
General Laborer	36	12.7%
Plumber or Pipe fitter	22	7.8%
Drywall	21	7.4%
Electrician	17	6.0%
Bricklayer or Mason	12	4.2%
Other	10	3.5%
Painter	8	2.8%
Heating, Ventilation, or Air Conditioning Installer	7	2.5%
Glass Worker or Glazier	2	0.7%
Roofer	1	0.4%
Insulation	1	0.4%
Total	283	100.0%

Some of these workers had also worked in other trades in their (usually brief) tenure in construction work. Table 6 shows the incidence of secondary trades, from most frequently cited to least.

Table 6 Secondary Trades of Respondents

Secondary Trade	# of Times Mentioned
General Laborer	21
Bricklayer or Cement Mason	17
Carpenter	17
Drywall (sheetrock) hanger	13
Ironworker	11
Electrician	9
Painter	8
Plumber of Pipefitter	4
Heavy Equipment Operator	3
Sheet Metal Worker	3
Painter	3
Other, or unclear answer	3
Air Conditioning Worker	2
Carpet Layer	1
Insulation	1

Eighty-one of the 283 respondents (28.6%) were union members. Of these eighty-one, fifty-two were members of either the Carpenters union (30) or the Ironworkers union (22). Other unions were the Plumbers and Pipefitters (11), the Laborers (9), the Electrical Workers (7), and the Bricklayers (1). One respondent who claimed union membership gave no name for his union. Average length of union membership was exactly three years to the month, with a range from two weeks to nineteen years. Most of these are at the low end of the spectrum: well over half had been union members for less than two years, with almost a third less than a year. Table 7 shows the spread.

Table 7 Length of union membership for union member respondents

Length of Union Membership	Number	Percent
Less than one year	26	32.1%
One year	18	22.2%
Two years	7	8.6%
Three years	8	9.9%
Four years	3	3.7%
Five years	2	2.5%
Six to Ten years	13	16.0%
Ten year and over	3* (10, 19, 25)	3.7%
No answer	1	1.2%
Total	81	100.0%

*The three respondents had ten, nineteen, and twenty-five years of union membership.

Almost a third (32.2%) personally earned less than $20,000 per year, and over fifty-eight percent earned less than $30,000. Table 8 shows a breakdown:

Table 8 Personal Yearly Income of Respondent Immigrant Construction Workers

INCOME RANGE	Number	Percent
Under $10,000	23	8.1%
$10,000 to $15,000	33	11.7%
$15,000 to $20,000	35	12.4%
$20,000 to $25,000	36	12.7%
$25,000 to $30,000	37	13.1%
$30,000 to $35,000	32	11.3%
$35,000 to $40,000	21	7.4%
$40,000 or more	34	12.0%
Wouldn’t answer; or gave unusable information	32	11.3%
TOTAL	283	100.0%

The respondents’ family income was generally higher than personal income; nevertheless, fifty percent had a family income below $30,000 per year, and thirty percent had a family income below $20,000 per year. (Average family income during Summer 2006 in the state of Florida was approximately $55,000 per year.) Table 9 shows the immigrant construction worker family income spread.

Table 9 Family Yearly Income of Respondent Immigrant Construction Workers

Income Range	Number	Percent
Less than 20,000	85	30%
20,000-29,999	58	20%
30,000-44,999	62	22%
45,000-59,999	22	8%
More than 60,000	29	10%
Wouldn't answer	27	10%
Total	283	100%

Forty-four percent had not completed high school or earned an equivalent diploma; yet the other end of the educational spectrum was also well represented. Twenty-one percent had taken at least some college courses, and thirteen percent had a college or graduate degree. Table 10 shows the schooling attainments of the immigrant respondents.

Table 10 Schooling Attainment of Respondent Immigrant Construction Workers

Amount of Schooling	Number	Percent
Less than high school	57	20%
Some high school (9th - 12th Grade)	61	22%
High school degree	98	35%
Vocational or technical school	7	2%
Some college (no degree)	22	8%
College or graduate degree	38	13%
Total	283	100%

Eighteen percent were U.S. citizens and almost three quarters were legally documented in one way or another. Twenty-seven percent were undocumented. Table 11 shows details.

Table 11 Legal Status of Respondent Immigrant Construction Workers

Legal Status	Number	Percent*
U.S. Citizen	50	18%
Not a Citizen; Documented	158	56%
Not a Citizen; Undocumented	75	27%
Total	283	100%

*Percentages do not add up to 100% because of rounding.

How representative is this sample of the overall population of immigrant construction workers in the area? The sample departs from our best estimate of the immigrant construction labor force in at least one important way. Union members were intentionally over-sampled to ensure a large enough group of union workers to make meaningful comparisons between union and non-union workers.

Beyond this over-sampling, the survey was not a random sample, as already noted, even though an attempt was made to approximate randomness as closely as possible. Therefore statistical tests of significance will not be used directly in the following analyses. Limited sample size due to resource limitations also means that some sub-sets of the data are too small for meaningful comparisons. Construction sites sampled were large or medium large, leaving out smaller commercial and residential construction workers. Despite these limitations, the database obtained in this research is still the largest and most representative sample of South Florida immigrant construction workers in existence, to the best knowledge of the author. It provides relatively good and extensive evidence that can be accepted as generally representative until better evidence is obtained.

EMPIRICAL RESULTS FROM SURVEY RESPONSES

The survey asks questions concerning six topic areas: (1) safety and health training received; (2) use of personal protective equipment on the job; (3) safety policies and practices of employers; (4) injuries and illnesses and related issues regarding workers compensation and disability; (5) other employer characteristics and practices which may be related to their safety practices; and (6) respondents’ evaluation of their employers’ attitude toward safety. This section will report results from responses by immigrant workers in each of these areas sequentially.

SAFETY AND HEALTH TRAINING RECEIVED

Immigrant respondents were asked which if any of six types of safety training they had received:

• the “OSHA 10 Hour Training,” which is a basic ten hour class offered by the Occupational Safety and Health Administration (OSHA) on safety and health matters;
• Scaffold Safety Training;
• CPR/First Aid Training in the past three years (A three year period was used because CPR certification expires after three years);
• Asbestos Awareness Training in the past three years (A three year period was used because asbestos awareness certification expires after three years);
• Hazardous Materials/Hazardous Location Training in the past three years (A three year period was used because Hazardous Materials/Location certification expires after three years); and
• Any other safety and health training.

OSHA 10-hour training and Scaffold Safety training had been received by seventy percent and sixty-eight percent respectively. Hazardous materials/area training was also received by over half of the respondents. Other types of training had been given to only one-fifth to one-third of the respondents. Table 12 shows the details.

Table 12 Numbers and Percentages of Respondents Receiving Various Types of Training

TYPE OF TRAINING	# who received training	% YES
OSHA 10-hr. Training	198	70%
Scaffold Training	192	68%
CPR/First Aid Training (last 3 yrs.)	97	34%
Asbestos Awareness Training (last 3 yrs.)	58	20%
Hazardous Material sTraining (last 3 yrs.)	160	57%
Other Safety Training	81	29%

Respondents were also asked about the language used to do the training, and their level of understanding of the training. In all but one case, between seventy percent and eighty percent had received training either in their native language, or with translation. In all cases, only one percent or two percent claimed they could not fully understand the training they had received. Table 13 shows details. (In this and following tables, “other training” is omitted, as it turned out to mean such different things that non-uniformity made the data meaningless.)

Table 13 Language and Level of Understanding of Training Received, by Type of Training

Type of Training	In Native Language	In English without Translation	In English with Translation	Fully Understood	Not Fully Understood
OSHA 10-hr. Training	69% (137)	30% (59)	1% (2)	99% (197)	1% (1)
Scaffold Training	73% (140)	26% (50)	1% (2)	98% (189)	2% (3)
CPR/First Aid Training	74% (72)	23% (22)	3% (3)	97% (94)	3% (3)
Asbestos Training	59% (34)	40% (23)	2% (1)	98% (57)	2% (1)
Hazardous Training	74% (119)	24% (39)	2% (3)	99% (159)	1% (2)

A separate question asked who provided the training. In virtually all cases, either the employer or the union (if there was one) was the provider. Between twenty-eight and twenty-nine percent of the immigrant respondents were union members, and the various unions were responsible for between fourteen and twenty-two percent of the training received by the entire sample. Especially for OSHA 10-hour training and Asbestos training, union members usually got their training from a union. Table 14 has details.

Table 14 Provision of Training by Unions, Employers, and “Other”

TYPE OF TRAINING	UNION PROVIDED (% AND #)		EMPLOYER PROVIDED	PROVIDED BY “OTHER”
	apprenticeship	not apprenticeship	(% AND #)	(% AND #)
OSHA 10-hr. Training	16% (31)	6% (12)	76% (151)	2% (4)
Scaffold Training	13% (4)	3% (5)	82% (158)	3% (5)
CPR/First Aid Training (last 3 yrs.)	13% (13)	3% (3)	74% (72)	9% (9)
Asbestos Training (last 3 yrs.)	14% (8)	7% (4)	74% (43)	5% (3)
Hazardous Training (last 3 yrs.)	12% (19)	2% (3)	82% (132)	4% (7)

Because the OSHA 10-hour training is basic training that all construction workers should have received immediately upon beginning work in the industry, immigrant respondents were asked how soon they received it after beginning work in construction. Answers ranged all the way from “before I started work” to “twenty two years.” The “average” time, inflated by some “outliers” who received training only after many years in the industry, was a little over a year and a half. The much more meaningful median (half longer, half shorter) was fourteen days, and almost one-third (65 of the the 198) had received their training within a day or less.

Most immigrant respondents who had received OSHA 10-hr. training had been asked to sign a statement acknowledging having received it; one 173 of the 198reported having signed such a statement. The same is true for the other types of training: scaffold safety training (149 out of 192); CPR/first aid training in the last three years (84 out of 97); asbestos awareness training (50 out of 58); and hazardous materials/areas training (134 out of 160).

The types of training requiring periodic (three year) re-certification (CPR/first aid, asbestos awareness, and hazardous materials/areas training) had been taken multiple times by some of the immigrant respondents. For each type of training, the most frequent response was only one training, but in all cases a majority had taken the training two or more times. Table 15 shows the details.

Table 15 Number of Times 3-year Certification Training Received, by Type of Training

Number of times received	CPR/First Aid Training	Asbestos Awareness Training	Hazardous Materials/Spaces Training
Once	42	22	51
Twice	27	21	37
Three times	12	5	33
Four times	4	5	13
Five times	5	0	9
Six times	2	1	1
Seven times	1	0	1
Eight times	1	1	0
Nine times	0	1	0
Ten times	1		6
12 times			2
20 times			3
24 times			1
30 times			1

Immigrant respondents were also asked how long their longest training in each of the certifiable areas had been. Responses ranged from less than one hour to more than forty hours. “Training” of less than one hour probably should not be counted as genuine training because its brevity makes it too superficial. We will return to this issue later, when analyzing relationships of training with other variables. Table 16 shows the range of responses regarding longevity of longest training in each area.

Table 16 Length of Longest Training, by Type of Training

Longest Training	CPR/First Aid Training	Asbestos Awareness Training	Hazardous Materials/Spaces Training
Less than 1 hour	7	5	16
1 hour	13	8	25
1.5 hours	0	3	6
2 hours	14	10	33
2.5 hours	0	0	2
3 hours	12	4	16
3.5 hours	1	0	1
4 hours	13	8	18
4.5 hours	1	0	0
5 – 9 hours	18	11	23
10 hours	8	5	10
11 – 15 hours	1	1	3
16 – 25 hours	6	0	4
26 – 40 hours	2	1	5
41 – 60 hours	1	2	0
Over 60 hours	0	0	1 (90 hours)

OTHER SAFETY AND HEALTH TRAINING

Asked to describe the type of training received, respondents displayed an enormous variation in what they considered “training.” Two of the eighty-one claiming some “other training” described weekly or monthly general safety meetings, not training sessions. The most common types mentioned were “general safety training” (twenty-one mentions), safety harness training (eleven mentions), fall protection training (eleven mentions), crane or crane rigging training (six mentions), “safety tools” training (five mentions), training videos (four mentions), and nail gun safety training (two mentions). Others mentioned once include: electrical grounding training, elevator safety training, deck safety training, safety flagging training, material safety data sheet (MSDS) training, fire safety training, ironlift operator training, safety glasses training, fork lift training, “more OSHA” training, “personal training,” etc. The responses to this question were too varied to provide much beyond a listing of training types.

USE OF PERSONAL PROTECTIVE EQUIPMENT

Respondents were asked to mark whether they “never,” “sometimes,” “regularly,” or “always” used various types of personal protective equipment on the construction job site. Table 17 shows the percentages and numbers for each response for seven types of protective equipment.

Table 17 Number and Percentage of Immigrant Respondents Using Various Types of Protective Equipment on the Job

TYPE OF PROTECTIVE EQUIPMENT	NEVER USE	SOMETIMES USE	REGULARLY USE	ALWAYS USE
Wear Work Boots	0.4%	1.1%	2.5%	96.1%
	1	3	7	272
Wear a Hard Hat	0%	0.7%	0.4%	98.9%
	0	2	1	280
Wear Work Gloves	8.5%	20.9%	12.4%	58.2%
	24	59	35	164
Wear Protective Eyewear	1.8%	8.5%	7.4%	82.3%
	5	24	21	233
Use Guards on Cutting Tools	10.8%	11.6%	7.8%	69.8%
	29	31	21	187
Use Hearing Protection	25.4%	31.8%	13.2%	29.6%
	71	89	37	83
Use Respiratory Protection	23.4%	33.1%	12.9%	30.6%
	65	92	36	85

Combining “regularly use” with “always use” to signify consistent use of these types of protective equipment, and combining “never use” and “sometimes use” to signify either no use or inconsistent use, one obtains the following results for each type of equipment:

• Wearing Work Boots: 98.6% consistently do; 1.4% do not
• Wearing a Hard Hat: 99.3% consistently do; 0.7% do not
• Wearing Work Gloves: 70.6% consistently do; 29.4% do not
• Wearing Protective Eyewear: 89.7% consistently do; 10.3% do not
• Using Cutting Tool Guards: 77.6% consistently do; 22.4% do not
• Using Hearing Protection: 42.8% consistently do; 57.2% do not
• Using Respiratory Protection: 43.5% consistently do; 56.5% do not

SAFETY POLICIES AND PRACTICES OF EMPLOYERS

The survey also asked about nine different employer safety policies and practices. Responses are summarized in Table 18.

Table 18 Percentages and Numbers of Immigrant Respondents Exposed to Various Employer Safety Policies and Practices

EMPLOYER PRACTICE	YES	NO	NO ANSWER OR EQUIVOCAL ANSWER
Weekly Safety Meeting	78% 222	22% 61	0 0%
Require Use of Body Harness	93% 263	4% 11	3% 9
Provision of Safety Program	77% 219	22% 63	0% 1
Provide Access to Material Safety Data Sheets (MSDS)	58% 163	38% 108	4% 12
Use of Ground Fault Electrical Outlets	83% 235	10% 28	7% 20
Use of Cut and Taped Electrical Cords	22% 62	77% 218	1% 3
Provision of Scaffold Hand Rails	87% 245	4% 11	10% 27
Provision of First Aid Kits	83% 236	16% 44	1% 3
Provision of Fresh Drinking Water	92% 260	8% 23	0% 0
Provision of Bathrooms	92% 260	8% 23	0% 0

For the 222 whose employer held safety meetings, one 175 (almost 79%) of the meetings were held in the worker’s native language, while forty-four (almost 20%) were in English and three (about one and a half percent) were in English and translated. All of the respondents claimed to fully understand the contents of the meetings.

INJURIES, ILLNESSES, AND RELATED WORKERS COMPENSATION AND DISABILITY ISSUES

The survey also asked about injuries, work-related illnesses, workers compensation, and disability payments. Results are briefly summarized in the following tables. Table 19 reports on injury or work-related illnesses in the past three years.

Table 19 Percentage and Number of Immigrant Respondents Experiencing a Workplace Injury/Illness in Past 3 Years; Those Requiring Medical Attention from Same; and Those Losing Work Because of Same

CONDITION	YES	NO
Had Workplace Injury or Work-Related Illness in Past 3 Years	12% 34	88% 249
Had Workplace Injury or Work-Related Illness in Past 3 Years that Required Medical Attention	10% 28	90% 255
Had Workplace Injury or Work-Related Illness in Past 3 Years that Caused Day or More of Lost Work Time	9% 26	91% 257

Of the thirty-four who had lost work time due to workplace injury/illness, thirty (10.6% of the entire immigrant sample) had lost time due to an injury. Of these thirty, eighteen had experienced this only once, nine had experienced this twice, one had experienced it three times, and two had experienced it six times. This totals to fifty-one instances divided among thirty workers, making for an average of a little over one and a half instances for those experiencing lost time injuries. The total amount of time lost varied widely, from one day to one hundred eighty days. (Unusable responses include two who claimed no time lost and one who did not answer.) Most immigrant respondents who had lost work days due to injury (16 of 27 usable responses) reported a total of three days or less lost lost, and only one reported an injury that was serious enough to require prolonged absence from work: 180 days. In total, 310 days of work were lost. Averaged over the entire sample of immigrant respondents, this comes to a little over one day lost per respondent in a three year period, or less than half a day lost per year. Table 20 shows summary figures.

Table 20 Three Year Injury Statistics for the Sample Immigrant Population

SEVERE INJURY CAUSING LOSS OF WORK DAY	NUMBER OF TIMES INJURED CAUSING LOSS OF WORK DAY	NUMBER OF DAYS LOST DUE TO WORKSITE INJURY	LOST DAYS DIVIDED BY NUMBER OF RESPONDENTS IN SAMPLE	AVERAGE ANNUAL LOST DAYS PER RESPONDENT IN SAMPLE
11% (30)	51	310	1.1	0.4

Respondents who had been injured on the job were asked if they had reported it. Of the thirty-three usable answers, twenty-seven reported that they had. The six who had not were asked why they had not. Three gave evasive answers like “I don’t know” or “I forgot” or chose not to answer. Two indicated that they did not think the injury serious enough to merit informing the boss (“not serious” and “I considered it not severe”). One stated that he was “reluctant to tell boss because of fear of missing work.”

The twenty-seven who had reported their injury were asked what had happened after they reported it. Field notes from the surveyors indicate that in the vast majority of cases, medical treatment of one sort or another was the outcome. Seven were sent to a medical clinic; five to the hospital (three of the five mention the emergency room); two were sent to a doctor; two received unspecified “medical attention;” one got a “test, X-ray;” and four were given on-the-spot first aid of some sort. One stated that he was given (unspecified) compensation. Five indicated treatment that could be interpreted as less helpful or friendly: three were just “sent home” while one went to the hospital emergency room on his own, and another’s treatment consisted of “drink water and rest.”

Only nine respondents (3% of the overall immigrant sample) had lost work time in the past three years due to a work-related illness (not injury). Most had experienced this only once. A total of twenty-nine days was lost. Averaged over the entire sample of immigrant respondents, this comes to approximately one-tenth of a day lost per respondent in a three year period, or miniscule three one-hundredths of a day lost per year. Table 21 shows summary figures.

Table 21 Three Year Work Related Illness Statistics for the Sample Population

SEVERE ILLNESS CAUSING LOSS OF WORK DAY	NUMBER OF TIMES ILLNESS CAUSES LOSS OF WORK DAY	NUMBER OF DAYS LOST DUE TO WORK RELATED ILLNESS	LOST DAYS DIVIDED BY NUMBER OF RESPONDENTS IN SAMPLE	AVERAGE ANNUAL LOST DAYS PER RESPONDENT IN SAMPLE
3% (9)	14	29	.10	.03

WORKERS COMPENSATION ISSUES

Most immigrant respondents had not filed a workers compensation claim or received any payment from the workers compensation system. There is some discrepancy in the figures, as only nine stated that they had filed a workers compensation claim in the past three years, yet thirteen claimed to have received a workers compensation payment for work performed in those same three years. Of the nine who said they had filed a claim, eight stated that they had received payment; thus the thirteen receiving payment included five who state that they never filed a claim. Probably those five had a claim filed for them (although the “filed a claim?” question explicitly asked about others filing for them); in any case, the figures are reported here exactly as they were given to the surveyors.

All nine applicants asked for medical expenses; seven asked for lost work time payments, and one asked for permanent disability. Four of the thirteen who said they had received a workers compensation payment did not state how much they had received. Payment to the nine who answered ranged between $120 and $30,000. All but two of the payments were under $2,000. The mean, or average payment was $4,750; the median (half more, half less) was $900. Table 22 gives a summary of filing and payment statistics.

Table 22 Workers Compensation Experiences of Immigrant Respondents Who Filed in the Past Three Years

FILED A CLAIM	FILED FOR MEDICAL EXPENSES	FILED FOR LOST WORK TIME	FILED FOR PERMANENT DISABILITY	RECEIVED W.C. PAYMENT	AVERAGE AMOUNT OF PAYMENT
3% (9)	9	7	1	13	$4750

The 274 respondents who had not filed a workers compensation claim were asked if their employer paid into the workers compensation system. Only 244 answered the question, making for an eighteen percent nonresponse rate. Of the remaining eighty-two percent, seventy-six percent indicated that they were covered and six percent that they weren’t or did not know. It is difficult to interpret the high nonresponse rate, but if it is added to the numbers of those who don’t have or don’t know if they have coverage, up to twenty-four percent could be without workers compensation coverage.

Only four respondents (1.4%) had ever been asked to sign a waiver of workers compensation coverage. Indicating that the experience was not a product of working for a tiny “fly by night” contractor, all four indicated that the employer making this request employed more than ten employees. Table 23 gives summary statistics concerning workers compensation.

Table 23 Workers Compensation Experiences of Immigrant Respondents Who Did Not File in the Past Three Years

HAVE COVERAGE	DON’T HAVE COVERAGE, OR DON’T KNOW	DIDN’T RESPOND ABOUT COVERAGE	ASKED FOR WORKERS COMPENSATION WAIVER
76% (185)	6% (15)	18% (44)	4 (employers employ >10 employees)

SELF ASSESSMENT OF HEALTH

Over a third of the immigrant respondents rated their own health as “excellent” and over half assessed it as either excellent or very good. Less than five percent rated their own health simply “fair,” and none rated themselves as “poor.” Table 24 shows results.

Table 24 Immigrant Respondents’ Self-Assessment of their own Health.

EXCELLENT	VERY GOOD	GOOD	FAIR	POOR
35.3% (100)	20.5% (58)	39.6% (112)	4.6% (13)	0% (0)

A large majority thought their health had not changed appreciably in the past year. Almost seventy-one percent compared their present health with that of one year ago as “about the same”; and deviations from that rating tended to move in the direction of improvement. Table 25 shows details.

Table 25 Respondents’ Assessment of Change in Their Own Health, Past Year

MUCH BETTER	SOMEWHAT BETTER	ABOUT THE SAME	SOMEWHAT WORSE	MUCH WORSE
12.8% (36)	11.7% (33)	70.8% (199)	3.9% (11)	0.7% (2)

SERIOUS INJURIES AND DEATHS AT WORK SITES

Immigrant respondents were asked if they had been working at a job site in the last year when a construction worker at the same site had to be taken to a hospital because of an injury. Thirty-nine percent (110) responded that they had. The total number of such incidents witnessed was 246, meaning that the “typical” witness had seen this a little over twice a year. Respondents were also asked if they had worked on a site since they started working construction when a construction worker died in a work related accident. Eighteen percent (50) had. Table 26 shows details.

Table 26 Percentage and Number of Immigrant Respondents Witnessing Serious Accident Requiring Hospitalization in Past Year, and Witnessing Accidental Death at Work Site in Entire Time Working in Construction

	ACCIDENT REQUIRING HOSPITALIZATION (PAST YEAR)	ACCIDENT CAUSING DEATH (ENTIRE TIME WORKING IN CONSTRUCTION)
PERCENTAGE OF RESPONDENTS (#)	39% (110)	18% (50)
NUMBER OF INCIDENTS WITNESSED	246	Not asked.

OTHER EMPLOYER CHARACTERISTICS AND PRACTICES THAT MAY BE RELATED TO THEIR SAFETY PRACTICES

The survey also asked a number of other questions concerning employers and the relationships of the immigrant respondents with them. The information solicited was thought to be possibly related to employers’ safety and health practices – for example, worse treatment in other respects may coincide with requiring employees to work in a less safe manner. Results will be briefly summarized here.

LENGTH OF TIME WITH CURRENT EMPLOYER

Respondents were asked how long they had been with their current employer. Over half had worked for their current employer less than a year, and almost seventy percent had less than two years in with their current employer. Table 27 shows details.

Table 27 Length of Time Immigrant Respondents had worked for their Current Employer

LESS THAN ONE MONTH	ONE MONTH TO LESS THAN ONE YEAR	ONE YEAR TO LESS THAN TWO YEARS	TWO YEARS TO LESS THAN FIVE YEARS	FIVE YEARS TO LESS THAN TEN YEARS	TEN YEARS OR MORE	NO ANSWER
6% (17)	48.4% (137)	15.5% (44)	18.7% (53)	6.7% (19)	3.9% (11)	0.7% (2)

NUMBER OF EMPLOYERS IN THE PAST YEAR

Most immigrant respondents had worked for only one construction employer in the past twelve months, and over eighty percent had worked for two or one. Table 28 shows details.

Table 28 Immigrant Respondents’ Number of Construction Employers in the Past Twelve Months

ONE	TWO	THREE	FOUR	FIVE	SIX	SEVEN	TEN	ABOVE TEN	NO ANSWER
57.6% (163)	23.7% (67)	9.9% (28)	4.6% (13)	1% (3)	0.35% (1)	0.35% (1)	0.35% (1)	0.71% (2)	1% (3)

HOW CURRENT JOB WAS FOUND

Over half of immigrant respondents got their job either through “word of mouth” in general or through referral by a friend or family member. Other methods categorized as “other” such as a labor pool or temp agency referral (eleven cases) or simply walking onto a job site (forty seven cases) were also frequent. Table 29 has details.

Table 29 Numbers and Percentages of Respondents Who got their Job in Various Ways

Want ad in paper	4% (10)
Word of mouth	7% (21)
Friend or family member	47% (132)
Union hiring hall	9% (26)
Referred by prior employer	5% (15)
Moved with employer from previous job	5% (15)
Other (walked on job site, temp agency or labor pool referral, radio, internet, etc.)	23% (65)

TYPE OF FIRM WORKED FOR

Respondents were asked if they worked for a construction firm, a temp help firm, or “other”. Ninety percent worked for a construction company (contractor or sub-contractor), while nine percent worked for a temp help firm and one percent (three people) worked for “other.” The three “other” respondents reported that they worked for a “straw boss,” a term whose meaning is not entirely clear. But it appears that a “straw boss” is equivalent to an extremely small sub-contractor who delivers workers to a firm and pays them out of his own pocket (whether legally or as part of the underground economy is not clear) after collecting a fee from the construction contractor. All three reported being paid by the “straw boss,” not the construction firm. Two reported working for their straw boss a half year and a year respectively; the third did not answer this question. Two reported preferring their current arrangement to working directly for the construction firm; the third was unsure of his preferences.

Table 30 Type of Firm Currently Working For

Construction firm	90% (255)
Temp help firm	9% (26)
Other	1% (3)

Of the twenty-six working for a temp help firm, over half (fourteen) had worked for this firm less than a year, and all had five years or less with the firm. All but two received their paycheck from the temp help firm rather than the construction firm. By a margin of fifteen to ten (with one not answering), these temp help employees would have preferred to get paid by the construction firm but were stuck with the temp firm for one reason or another.

EMPLOYEES ON CURRENT JOB SITE AND TOTAL EMPLOYMENT OF CURRENT EMPLOYER

Employment at the immigrant respondents’ current job sites ranged from two to one thousand. The mean (average) was one hundred fifty six, while the median (half more, half less) was one hundred. In addition to working on rather large job sites for this industry, the immigrant respondents also tended to work for much larger than average employers. Sixty-six percent of them worked for an employer with one hundred employees or more, and more than a quarter had employers with five hundred or more employees. Details are in Table 31.

Table 31 Number of Employees at Current Job Site, and Total Employment of Employer*

RANGE	NUMBER OF EMPLOYEES AT CURRENT JOB SITE	TOTAL EMPLOYMENT OF EMPLOYER
Less than 10	4% (10)	2% (7)
10-14	9% (24)	5% (14%)
25-99	33% (93)	20% (57)
100-499	46% (128)	40% (113)
500-999	9% (24)	26% (74)
Don’t know	0 (0%)	6% (16)

*Numbers do not always add up to 283 due to a few nonresponses to each question.

UNIONIZATION STATUS OF CURRENT EMPLOYER

Forty-five percent of the immigrant respondents stated that their employer was completely non-union; twenty percent stated that it was completely unionized. Most of the rest indicated some portion of the employer’s workforce, but not all, was unionized. The unions mentioned most frequently as representing the employer’s workers were the Carpenters, the Ironworkers, the Electricians, the Plumbers, and the Laborers. Table 32 summarizes unionization status.

Table 32 Immigrant Respondent Assessments of How Unionized Employers Are

ALL EMPLOYEES UNION	MOST EMPLOYEES UNION	SOME EMPLOYEES UNION	NO EMPLOYEES UNION	DON’T KNOW
20% (56)	11% (32)	22% (61)	45% (126)	3% (8)

AVERAGE DAYS WORKED PER WEEK IN CONSTRUCTION IN PAST YEAR

On average, respondents averaged 5.36 days of construction work per week, while working in construction. Over ninety-eight percent worked either five or six days a week. The mean number of hours worked was 44.48; the median (half more, half less) was 40 hours. Table 33 has details.

Table 33 Average Days Worked per Week and Average Hours Worked per Week in Past Year When Working in Construction

Average Days Worked per Week, While Working in Construction	3	0.7% (2)
(Average for all 283 Immigrant respondents is 5.36 days)	4	0.4% (1)
	5	61.8% (175)
	6	36.7% (104)
	7	0.4% (1)
Average Hours Worked per Week, While Working in Construction	Minimum: 24 hrs
(Average for 282 Immigrant respondents who answered is 44.48 hrs.)	Maximum: 70 hrs.
	Median: 40 hrs

TYPES OF PAYMENT AND RATES OF PAY

Eighteen percent (51) of the immigrant respondents indicated that at some point (not necessarily with the current employer) they had been paid for construction work in cash. Of the fifty-one who had, twenty-nine indicated that the employer employed more than ten workers and twenty-one indicated a small employer with less than ten employees. In virtually all cases (47 of the 51 cases) the employer who had done this was nonunion.

Far fewer had been illegally asked to sign a “1099 form” declaring themselves independent contractors even though they were working by the hour: eight percent (23 respondents). Of the twenty-three who had been asked, nineteen were asked by employers of more than ten workers and four were employers of less than ten. Eighteen of the twenty-three were nonunion employers. Table 34 gives details.

Table 34 Number of, and Characteristics of, Firms Paying Respondents in Cash or Requiring Dishonest Filling Out of Independent Contractor Form

EMPLOYER PRACTICE	YES	NO	KNOWN EMPLOYER CHARACTERISTICS FOR “YES” ANSWERS	NUMBER
Paid in Cash?	18% (51)	82% (232)	Less than 10 Workers	22
			More than 10 Workers	29
			Non-Union	47
Asked to Dishonestly Sign an Independent Contractor Form?	8% (23)	92% (260)	Less than 10 Workers	4
			More than 10 Workers	19
			Non-Union	18

All but eleven of the immigrant respondents indicated that they were usually paid by the hour. Five were paid by the day; four by the job (piece rate), and two by salary. Those working by the hour averaged $14.76 per hour, from a low of $5.15/hour to a high of $40.00/hour. Those paid by the day tended to make less; salaried were generally highly paid, while those paid by the job (piece rate) averaged about the same as hourly workers. Table 35 shows details.

Table 35 Type of Pay and Levels of Pay for Immigrant Respondents

	PAID BY THE HOUR	PAID BY THE DAY	PAID BY THE JOB	PAID BY SALARY
Percent (#)	96% (272)	2% (5)	1% (4)	1% (2)
Hourly Earnings	Average: $14.76	Average: $11.78/hr.	Average: $15.00/hr.	Average: $28.75/hr.
	Low: $5.15	Low: $10.00/hr.	Low: $10.00/hr.	Low: $25.00/hr.
	High: $40.00	High: $14.40/hr.	High: $25.00/hr.	High: $32.50/hr.
	Below $10: 11% (30)	$10.00 hourly: 40% (2)	$10.00 hourly: 25% (1)	$25.00 hourly: 50% (1)
	$10-$11.99 15% (41)	$12.00 hourly: 20% (1)	$12.00 hourly: 25% (1)	$32.50 hourly 50% (1)
	$12-$13.99 18% (49)	$12.50 hourly: 20% (1)	$13.00 hourly: 25% (1 )
	$14-$15.99 22% (59)	$14.40 hourly: 20% (1)	$25.00 hourly: 25% (1)
	$16-$17.99 10% (28)
	$18-$19.99 12% (33)
	$20-24.99 6% (17)
	$25 up 6% (15)

PROVISION OF A RETIREMENT OR SAVINGS PLAN

Thirty-five percent (98) of the immigrant respondents indicated that their employer offered a retirement or savings plan. Of those with a plan, sixty-seven percent indicated that the employer contributed to it. Even though unionized respondents comprised less than thirty percent of the sample, sixty-nine percent of the retirement/savings plans were union, indicating the better retirement provisions available to union members. Table 36 provides details.

Table 36 Retirement or Savings Plan Provision and Types

PROVISION OF A HEALTH INSURANCE PLAN

Almost fifty-six percent (158) of immigrant respondents indicated that their employer provided a health insurance plan but only a quarter of those indicating such a plan were able to state what percentage of the health insurance premium was paid by the employer. Table 37 provides details.

Table 37 Number of Immigrant Respondents Offered Health Insurance Coverage, and Percentage of Insurance Premiums Paid by the Employer

	YES	NO	NO ANSWER
Offered Health Insurance Coverage?	55.8% (158)	43.1% (122)	1.1% (3)
Percentage of Premium Paid by the Employer	100%: 7% (11) 99-75%: 9% (14) 74-50%: 8% (12) < 50%: 2% (3) Don’t know: 75% (118)	NA	NA

PERCEPTIONS OF EMPLOYER ATTITUDES AND PRACTICES CONCERNING SAFETY

Immigrant respondents were asked to state if they “strongly agree”, “agree”, “disagree”, or “strongly disagree” with a series of statements that indicate their assessment of their employers’ attitudes and practices concerning safety. Overwhelmingly they felt that their employers were safety conscious, although sixty percent also stated that their work conditions were dangerous. Table 38 shows the percentages and numbers of each response for nine statements.

Table 38 Number and Percentage of Immigrant Responses Agreeing or Disagreeing with Evaluations of Employer Safety Attitudes and Practices

	STRONGLY AGREE	AGREE	DISAGREE	STRONGLY DISAGREE
My foreman is concerned about worker safety	45.2% (126)	52.3% (146)	2.2% (6)	0.4% (1)
My contractor (employer) is concerned about worker safety	45.4% (128)	49.6% (140)	4.3% (12)	0.7% (2)
Unions lead to safer jobs	37.3% (98)	52.1% (137)	9.1% (24)	1.5% (4)
My work conditions are dangerous	16.7% (47)	43.3% (122)	34.8% (98)	5.3% (15)
My work area is kept clean	32.9% (92)	56.4% (158)	10.4% (29)	0.4% (1)
My work area is cluttered	1.8% (5)	14.2% (40)	72.0% (203)	12.1% (34)
My job site has a good safety program	34.4% (96)	57.7% (161)	7.2% (20)	0.7% (2)
I have too much to do to be able to follow safe work practices	2.5% (7)	18.1% (51)	66.3% (187)	13.1% (37)
Where I work, productivity is more important than worker safety	2.5% (7)	16.0% (45)	60.9% (171)	20.6% (58)

If we combine “strongly agree” with “agree” to signify general agreement and “strongly disagree” with “disagree” to signify disagreement with these statements, we obtain the following results:

• Foremen is concerned about worker safety: 97% agree; 3% disagree;
• Employer is concerned about worker safety: 95% agree; 5% disagree;
• Unions lead to safer jobs: 89% agree; 11% disagree;
• My work conditions are dangerous: 60% agree; 40% disagree;
• My work area is kept clean: 89% agree; 11% disagree;
• My work area is cluttered: 16% agree; 84% disagree
• My job site has a good safety program: 92% agree; 8% disagree;
• I have too much to do to follow safe work practices: 21% agree; 79% disagree;
• Where I work, productivity is more important than worker safety: 18.5% agree; 80.5% disagree.

As a further test of respondent’s assessment of their employer’s commitment to safe policies and practices, respondents were asked whether they would report a safety violation to their employers if they were aware of it. Ninety-six percent said yes, a further indication of their confidence that the employer was serious about safety. Table 39 shows results.

Table 39 Willingness of Respondents to Report a Safety Violation

	YES	NO	UNSURE
Would You Report a Safety Violation?	96% (271)	3% (8)	1% (4)

The twelve who answered no or were unsure were asked why they would not or might not. The surveyors’ field notes about answers indicate that fear is a primary reason:

• “They would probably fire him”
• “Sometimes the employer can fire you”
• “He minds his own business”
• “Fear”
• “He is new, and does not know how to do it”
• “People may take it against him”
• “Someone else is in charge of that”
• “Someone else is in charge”
• “Work has to get done. (But, if safety inspector sees it, he will handle it.)”
• “He would avoid reporting it unless it was serious”
• “He would just call person’s attention to it”
• “Talk to guys directly; they have to leave if I tell them to.”

With the exception of the last response (which seems to be from a supervisor or safety person), these responses all seem to either explicitly or implicitly indicate fear or reticence because of possible negative consequences to them if they did report a safety violation. There is a high congruence between the expressions of fear (or reticence) stated above and the same respondent’s negative assessment of their foremen’s (and employer’s) concern with safety. All except one either disagreed or strongly disagreed with the statement that their foreman was concerned about safety, and all but two felt the same way concerning their employer. So, despite general belief that foremen and employers were concerned with safety, a small minority (four percent) felt intimidated and fearful.

¹ Of the 283 immigrant workers, 202 were non-union and 81 were union members.
² To obtain the raw data from this study on this finding or any other finding, please contact the author.