Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Addressing Multilingualism in Construction Workplace Education: Results of a Pilot Experiment

by James Wilkins, Marilyn Alibutod & Anindita Nugroho - 2013

Background: In the United States, foreign-born workers of Hispanic origin have been central to the construction industry for decades. Many of these workers have experienced only limited education in either English or Spanish and most are bilingual to some degree.

Purpose: As employers are beginning to recognise the benefits of providing access to the OSHA 10- and 30-hour Construction Safety Training Course, the industry has begun to experience a downward trend in preventable accidents, illnesses and fatalities. Despite this decline, a detailed review of recently published research indicates that foreign-born workers of Hispanic origin remain nearly 70% more likely to be involved in a work-related incident than their American-born counterparts. In order to develop and provide effective workplace learning programmes it is necessary to assess the impact which these workers’ literacy and confidence in a second language has on their professional capabilities.

Research Design: The authors of the present article examined the problem by utilizing a simple post-test only randomised experimental design to measure knowledge retention among this demographic immediately following the completion of OSHA 10-hour Construction Safety Training Courses when the languages of instruction were English-only, Spanish-only and both English and Spanish.

Conclusions: The data revealed that workers who fall into this category retained more knowledge from the training course when it was taught using a translanguaging approach. This data will be of value to designers of workplace education programmes, who will be better placed to incorporate a bilingual element into their instructional methodologies.


Industry leaders in construction are making progress towards implementing and maintaining a culture of safety awareness in the workplace. Modern-day vocational training efforts are largely responsible for the steady reduction in the number of serious accidents and fatalities recorded in the construction industry year on year; however, industry participation remains riskier for foreign-born workers of Hispanic origin. They are still nearly 70% more likely to be involved in an occupational accident that leads to physical or psychological injury than their American-born counterparts.  A significant target for advocates of safe working practice requires all participants in the workforce to have access to effective, life-saving health and safety training. For this to happen, it is important for industry leaders to understand what works in workplace education and development efforts and, similarly, what is detrimental to the process of making employees more aware of and better equipped to deal with the professional hazards they encounter.

In recognition of the significant contributions made by foreign-born workers of Hispanic origin, and their vulnerability, researchers developed an experiment that examined the use of an instructional technique that incorporated translanguaging in the delivery of an OSHA 10-hour Construction Safety Training Course. The present article is aimed at providing background information on translanguaging and to share the results of a pilot  experiment that examined the use of an instructional technique in the delivery of an OSHA 10-hour Construction Safety Training Course. Analysis of the exploratory study was encouraging, and suggested that translanguaging was a favourable implementation for improving knowledge retention among foreign-born workers of Hispanic origin. Also, as a result of the study, researchers were able to improve the experimental design prior to full-scale deployment.  


The United States construction industry employs the greatest proportion of Hispanic workers of any national industry (Walter, 2010). In 2010, construction firms and private clients employed 2.1 million Hispanic workers, which represented 24.4% of overall number of workers in construction (United States Bureau of Labour Statistics, 2011b; United States Census Bureau, 2011). The same demographic group also accounted for 23.2% of all fatalities in the construction industry (United States Bureau of Labour Statistics, 2011c). According to Walter (2008), foreign-born workers of Hispanic origin have “a 70% higher rate of work-related injury [or] death compared to native-born Hispanic workers.”  

In recent years, rulemaking pertaining to the provision of health and safety training in the construction workplace has been significantly strengthened. However, very little has been done to address the special needs of foreign-born workers of Hispanic origin in respect to transitional education and development. According to a study conducted by Wilkins (2009), of construction workers who had undergone and completed a 10-hour Construction Health and Safety Training Course, only 14% indicated that training was offered in a language other than English. While health and safety training reforms have shown great promise over the years, more progress is needed so that the provision of life-saving health and safety training can better reflect the landscape of the construction workforce.


With reference to the risk to life and livelihood faced by all employees working at the project level in construction, this study explored the efficacy of integrating translanguaging into the instruction design for the OSHA 10-hour Construction Safety Training Course offered to foreign-born workers of Hispanic origin. If organisation leaders can identify whether translanguaging is an effective method of reaching these workers – and the extent to which it makes a difference – then they will be in a better position to make effective strides toward incorporating the approach in workplace training efforts.  The purpose of this study is to describe any trends in the scores achieved by foreign-born workers of Hispanic origin on a knowledge quiz undertaken following completion of the OSHA 10-hour Construction Health and Safety Training Course.


Do foreign-born workers of Hispanic origin perform better on knowledge tests that follow the provision of the OSHA 10-hour Construction Health and Safety Training Course when translanguaging is incorporated into the instructional design, as opposed to when training is conducted wholly in English or Spanish?


Foreign-born workers of Hispanic origin perform better on knowledge tests that follow the provision of the OSHA 10-hour Construction Health and Safety Training Course when the training course, instructional materials and tests are presented wholly in the Spanish language.


 The concept underpinning this study is one based on rationality. Individuals who immigrate to nations where languages other than their mother tongue are spoken typically feel at ease and more comfortable when interacting in their native languages (Blank & Slipp, 1994). When workplace education is provided using a methodology that embraces their bilingualism, then there may be some improvement in the outcomes for workers. Critically examining performance in knowledge tests undertaken by foreign-born workers of Hispanic origin when training is provided in English, Spanish, and both languages will bring organisation leaders one step closer to designing learner-focused workforce education and development environments. An improvement in the rate of overall retention of knowledge gained from such learning environments will better enable workers to transfer knowledge to safer working practice, and will hopefully further reduce the incidences of fatal occupational accidents.


The construction industry boasts a diverse workforce. According to the Bureau of Labour Statistics (2012) an annual average of 7.2 million people worked the in construction and extraction occupations during 2011, with Hispanic workers accounting for approximately 30% of this total figure. According to the Centre for Construction Research and Training (2007), construction workers are, on average, younger than the national labour force, with an average age of 39 in 2005 (compared to an average age of 37.8 in 2007); however, the industry is experiencing an upward shift that is leading to an increase in older workers because fewer young workers are entering the workforce.

Projections anticipate a potential doubling of the number of workers over age 65 in the next few decades. The retirement of the baby boomers has created a shortage among construction workers. In the short term this shortage was compensated for by an influx of migrant labourers; however, many of these immigrants had a lower level of education than their American counterparts. On average, Hispanic workers are six years younger than other workers. Only 23% of non-Hispanic construction workers are under the age of 30 whereas 38% of Hispanic workers fall into that age bracket. Illiteracy and lack of educational attainment are significant concerns within this workforce (Crowley & Lutz, 2007; Centre for Construction Research and Training, 2007). The likelihood that workers of lower literacy will return to formal training (or, indeed, reliable and systematic training of any kind) in order to improve their English language skills is very low. Consequently, business and industry leaders are now facing up to the challenge to develop workplace education programming which responds directly to the needs of the worker.



Transformative learning experiences were investigated by Giles and Alderson (2008), who examined them by means of a literacy project. They emphasized the importance of creating learning environments within which all students felt a sense of inclusion and respect. Such environments can be created by encouraging students to help one another, alleviating the feeling of being judged, and encouraging students to seek help and to share their cultural experiences. Hearing others’ stories was found to strengthen the relationship between students; however, participants in that exercise were all native English speakers, and were therefore reasonably secure in their abilities to effectively communicate in English, thereby being understood by their fellow learners and facilitator. Foreign-born workers face a special challenge in this area, although it is reasonable to suppose that they also benefit from an opportunity to share their experiences in workplace training environments. By enabling such interaction it is possible to transform the thoughts of the students so as to reflect the knowledge transferred in the training (Bedi, 2004). This exercise improves the likelihood that the training will be more meaningful to the workplace learner and have a direct impact on safety practice on the construction site.  


Approaches to adult learning in the context of safety training were studied by Galbraith and Fouch (2007). Their assessment of adult learning theories included a review of sensory stimulation theory, cognitive theory, reinforcement theory, facilitation and andragogy, which they claim are the most common forms of mature student training and instruction. In particular, their study focused on andragogy. They identified three principles of andragogy that are applicable to safety training: needs assessment (including contribution from the participants concerning the content to be learned); safety of the environment with regard to learners feeling comfortable and confident in expressing their thoughts; and the relationship between the trainer and the trainees. They compared two training sessions which they refer to as representing the ‘original’ training practice and the ‘new’ training practice. In the original training, trainees appeared to assimilate knowledge in a classroom environment, but subsequently failed to apply it in their professional lives. The new training specifically addresses the connection between theory and practice, and sets clear learning objectives which are discussed and reviewed throughout by students. The original training did not specify objectives, placed limited emphasis on discussion, used clip art as a visual, and was lecture-based. The authors concluded that their new, more andragogic method, was more effective and that their positive impact was demonstrable among those who had completed such a course.

The new training presented learners with objectives from the beginning; discussion was emphasized; and visuals included actual photographs, and were arranged in sections that delivered “whole-part-whole information” in “bite-sized chunks” (Galbraith & Fouch, 2007, p. 39). Scenarios and problems presented in the new training simulated the actual work environment experienced by trainees. While on the course, students were observed at random in order to establish whether or not they were adopting behaviours such as wearing protective gear properly, implementing good housekeeping practices, and labelling hazardous waste containers properly. Under the new training, there was a significant decrease in the number of safety violations in the workplace. The limitation of this study, of course, was that neither national origin nor language background were taken into account. This must be the next step in recognising and addressing the practical needs of the construction workplace in 2012.


The term translanguaging is most widely used when referring to language learners and classroom practices. It is also linked closely with the term code-switching, although the two terms are uniquely distinct.  According to García (2009), translanguaging demonstrates different ways in which bilinguals construct meaning through the use of more than one language in the learning environment.  Translanguaging aids in communication but also allows the bilingual speaker to make deeper connections by associating content with pre-existing knowledge in any language.  

Often the process of translanguaging is most easily understood in the context of code-switching, which can be defined as “the process of going back and forth from one language to another” (García, 2009, p. 8).  This may happen inter-sententially after a phrase or sentence, as well as intra-sententially, within one sentence. Code-switching often happens spontaneously among bilingual speakers, although it can also be used to emphasize statements within conversations or even reinforce one’s cultural identity.  Since code-switching is used mostly to mark speech or clarify a concept, it can be considered a function of an individual’s process of translanguaging.  “Regardless of the speaker’s choice in switching between two languages, this act demonstrates the ability to respond to external cues and the rules of a given linguistic system” (García, 2009, p. 8).


Although most of the established learning environments of bilingual learners tend to separate language use to ensure that appropriate attention is given to the rules governing one language or another, education programmes in construction workplaces across the United States are typically offered in English-only formats.  Since code-switching or native language use is prohibited, the benefits of using translanguaging to aid in student engagement and comprehension are lost.  Many believe that exposure to and interaction in the target language will result in the more rapid acquisition of that language.  Harper and de Jong (2004) comment that “in many introductory English as a Second Language (ESL) workshops outside of the bilingual profession, the tendency is to over-simplify theory thereby over-emphasizing overlap between first and second language learning” (p. 153).  In other words this assumption claims that students who have a native language other than English (LOTE) acquire their second language in the same way that monolingual students acquire their first language.

The main deficiency in this theory is that it fails to acknowledge the specific needs of foreign-born workers of Hispanic origin – the subjects of the present study.  The way in which a child acquires his first language (L1) is completely different from the way that additional languages are attained. Mere exposure to the target language is largely insufficient to develop second language (L2) proficiency, especially for older workers who must construct and negotiate abstract concepts and complex language. Migrant professional labourers typically lack the time and facility to master a second language in this way. Transmuting theoretical words, images and phraseology into a practical lexicon for use in the workplace is a challenging and constraining exercise. The alienation brought about by this insistence on a purely immersive, English language only environment is potentially very hazardous, since “LOTE students often require conscious attention to the grammatical, morphological, and phonological aspects of the English language” (Harper & de Jong, 2004, p. 153).

In workplace learning environments, when foreign-born workers of Hispanic origin are explicitly denied translanguaging opportunities, they simply do not participate fully in any training regimen. It takes between two and three years to become proficient in basic communication skills in a second language and between four and ten years to attain competence in L2 academic skills (Lucas & Katz, 1994, p. 538). Furthermore, if non-native English speakers are immersed in their second language at the exclusion of their first, they will not have access to the content area knowledge and academic skills that their English-speaking peers are learning.  This may result in students falling further and further behind.  If content was discussed in the native language of trainees, students would be able to interact more effectively with the instructional process. When foreign-born workers are given access to learning materials in their native language, they are more likely to build on prior knowledge and create connections between L1 and L2 content. Those connections may be invaluable in a bilingual worker’s ability to translate knowledge gained during safety training initiatives into safe work practice.

Benefits of Incorporating the Native Language into Instruction

Using the native language within various programs has proven to be of significant benefit to learners. Native language use is not intended to replace or encourage using L1 in place of the L2 when the goal is to operate professionally in a second language; but rather to realize that, at times, it may be more appropriate to include content in a student’s native language in support of the main course content.  A study conducted by Storch and Wigglesworth (2003) concluded that when clear goals and parameters are set for allowing the use of one’s native language in certain contexts that freedom often helps workers remain focused on the task at hand. Eight out of twelve subjects confirmed that the use of the L1 helped to complete tasks more efficiently. As Lucas and Katz (1994) suggest, native language use can, first and foremost, increase students’ willingness to learn by reducing the extent to which the need to operate functionally using L2 is disorienting to them. The bulk of their research investigated Special Alternative Instructional Programs (SAIPs) where English Language Learners from various sites were observed in their classrooms. Across all the sites involved in the study, group work was a key teaching strategy.  Teachers set up situations or activities that specifically called for students to use the L1; for example when collaborating on group writing assignments.  Other strategies included pairing less fluent speakers with more fluent speakers, providing bilingual dictionaries. When appropriate, resources were provided in both L1 and L2 for non-native speakers.  Even in ESL programs and those which use English-only as the instructional method of choice, monolingual teachers can still create language safe environments where L1 is deemed a resource rather than a detriment. It is important that students are not led to feel culturally inferior because English is their second language. When taught in ways which allow them to access both their L1 and L2 abilities, they are likely to progress more rapidly and develop higher self-esteem which, in turn, will support their learning experience (Lucas & Katz, 1994, p. 559).  The decisions about when, how, and how much to incorporate the L1 into the classroom, however, depend on each individual programme and unique challenges it faces.


Quality instruction, which includes the native language when appropriate, the amount of exposure to L2 incorporated in the training programme, and use of meaningful language, significantly affects the degree of success in second language acquisition for foreign-born workers of Hispanic origin.  In the case of the 10-hour OSHA Construction Safety Training course, it would seem beneficial to offer classes either bilingually or with inbuilt L2 language support measures.  If a bilingual instructor is not readily available, language support could take the form of providing hand-outs with important notes written in the native language for L1 Spanish students; providing visuals when key terms are spoken in English; or using a large amount of modelling within the training environment. In this way native Spanish speaking workers will be more actively engaged and all workers will have access to the content presented.


This study utilized data from foreign-born construction workers of Hispanic origin who undertook an OSHA 10-hour Construction Safety Training Course at a company in New York City. For the purpose of this study, the training company randomly allocated workers with Spanish-speaking backgrounds who registered for safety training between June and August 2011 to either (1) a program delivered only in English (“English-only” group), (2) a program delivered only in Spanish (“Spanish-only” group), or (3) a program delivered in mix of English and Spanish (“Translanguaging” group).  

Following the provision of the training, workers were required to complete a short, 33-item quiz on the topics that were covered during the sessions. The quiz was administered in a language corresponding to the training, such that the English-only group completed a quiz in English, the Spanish-only group completed a quiz in Spanish, and the Translanguaging group received a quiz that presented questions both in English and Spanish.

As workers were randomly assigned to training groups, and no pre-test data was available, the study employed a simple post-test-only randomized experimental design. The study sample surveyed included twenty (20) workers randomly selected from each training group, with a total of sixty (60) workers.


Workers who completed a training program delivered in Spanish (“Spanish-only” group) performed better on the safety examination set after the training than those who completed a training program delivered in English (“English-only” group). Workers who completed a training program delivered in a mixture of English and Spanish (“Translanguaging” group) performed better than those in either of the other two groups. The mean of workers’ scores in each group is presented in Table 1, and suggests that the differences between the three groups’ scores are quite significant. Results of the analysis of variance (ANOVA) confirm that the differences of scores between the three groups are statistically significant at the 99% level (F=77.683, p<.001).

Table 1: Scores (percentage correct) on post-training test, by language of instruction group

Language of instruction

























A visual depiction of the mean scores of workers in each language group is presented in Figure 1, with error bars denoting standard deviation. This graph illustrates that the average score of workers in the Spanish-only training group was 1.2 standard deviations higher than workers in the English-only group. The average scores of workers in the Translanguaging group was more than 2 standard deviations higher than workers assigned to the English-only group, and more than 0.75 standard deviations higher than workers assigned to the Spanish-only group.

Figure 1. Visual depiction of the mean scores achieved by workers in each language group (%)



Although it was predicted that those workers who received instruction in Spanish would perform more successfully, the results show that the group which used translanguaging techniques ultimately achieved the highest scores. This finding supports the notion that there is an apparent benefit for students in learning purposefully in both their native and second languages. As suspected, the group which was instructed solely in English performed least well.  Denying bilingual learners the opportunity to conduct some or all of their training in their native tongue has a detrimental effect, since it obliges them to engage in a learning process without access to the tools which are most efficient for them. It is significant to note that members of the group who were instructed only in Spanish scored lower than those in the Translanguaging group.  This is of particular interest since it would seem that the workers would be more comfortable and more readily equipped to undergo training and complete the quiz in their native language.  As the research has shown, translanguaging is not intended to replace or encourage using the L1 in place of the L2; but rather to realize that, at times, it may be more appropriate to incorporate elements of tuition in a learner's native language in carefully determined ways. A common thread across the research is that during group work, learners who were permitted to use their native language were able to clarify and focus on the task at hand more easily.


This study revealed that when dealing with foreign-born workers of Hispanic origin, the incorporation and recognition of their native language into training courses has a positive effect on their overall comprehension of material. Wilkins (2009) provided that greater scores on the same measure utilised in the present study was associated with better safety practices, with attendant benefits in reducing the incidence of accidents and safety violations on construction sites. Translanguaging offered these workers a chance to navigate information presented in their second language by using prior knowledge obtained from the use of their native language. Proper incorporation of the students’ L1 during training results in students being better able to understand the task at hand and master it more thoroughly, displaying a deeper knowledge of the instructional material under consideration.  Furthermore, including the use of students’ native language reinforces high self-esteem and strengthens identity among workers who identify both with fellow linguists and with the organization which acknowledges their linguistic background.  In other words, providing bilingual support in the workplace; whether in the form of bilingual trainers or the dissemination of targeted, multilingual instructional materials, has a net positive impact on the contribution which non-native English speakers are able to make.  This democratisation of information and training supports a sense of equality within the workforce and improves the prospects of all who complete OSHA training courses, saving time and money for employers and reducing the risk to themselves.


A number of lessons from this study are vital to preparing the experiment for full-scale deployment. The present study focused exclusively on the trainees and the researchers later realised that more details on the trainer’s proficiency in both languages were needed to determine the possible effect it may have had on learners being able to understand the words, phrases and colloquialisms used during the training sessions. Further to that, prior research has established that trainees perform better in terms of knowledge transfer when facilitators of the 10-Hour Construction Safety Training Course are perceived to be of a higher calibre by trainees (Wilkins, 2011). The ability to strongly connect with trainees in through language is certainly a way to enhance perceptions of quality and competence, since one must first be understood before being qualitatively evaluated. In order to more fully understand this phenomenon, further investigation regarding how proficient the workers are in their native language may be necessary.  Several factors, such as years of formal education in the native language, amount of time spent in the United States, and exposure to the second language will necessarily affect these findings.

In the full-scale study, researchers will more clearly identify a population of concern and an appropriate sample size calculation to ensure that the respondents adequately represent the study population.


Blank, R., & Slipp, S. (1994). Voices of diversity. New York, NY: AMACOM.

Garca, O. (2009). Bilingualism and translanguaging. In Bilingual education in the 21st century: A global perspective (pp. 42-72). Malden, MA: Wiley-Blackwell.

Harper, C., & de Jong, E. (2004). Misconceptions about teaching English-language learners. Journal of Adolescent and Adult Literacy, 48(2), 152-162. doi:10.1598/JAAL.48.2.6

Lucas, T., & Katz, A. (1994). Reframing the debate: The roles of native languages in English-only programs for language minority students. TESOL Quarterly, 28(3), 537-561. doi:10.2307/3587307

Storch, N., & Wigglesworth, G. (2003). Is there a role for the use of the L1 in an L2 setting? TESOL Quarterly, 37(4), 760-770.

United States Bureau of Labour Statistics. (2011a). Employed persons by occupation, race, Hispanic or Latino ethnicity, and sex. In Household data: Annual averages (sec. 10). Retrieved November 2, 2011, from United States Department of Labour website: http://www.bls.gov/cps/cpsaat10.pdf

United States Bureau of Labour Statistics. (2011b). Employed persons by occupation, race, Hispanic or Latino ethnicity, and sex. In Household data: Annual averages (sec. 13). Retrieved November 2, 2011, from United States Department of Labour website: http://www.bls.gov/cps/cpsaat13.pdf

United States Bureau of Labour Statistics. (2011c). Fatal occupational injuries by selected worker characteristics and selected industry [Data file]. Retrieved from http://data.bls.gov/gqt/RequestData

United States Census Bureau. (2011). Labour force, employment and earnings. In Occupations of the employed by selected characteristic: 2010 (table 619). Retrieved from http://www.census.gov/compendia/statab/2012/tables/12s0620.pdf

Walter, L. (2008). Hispanic workers face higher fatality rates. EHS Today, safety. Retrieved from http://ehstoday.com/safety/ehs_imp_80766/

Walter, L. (2010). Hispanic construction workers face health care challenges. EHS Today, safety. Retrieved from http://ehstoday.com/construction/news/hispanic-construction-workers-health-care-challenges-8796/

Wilkins, J. R. (2009). Variables affecting employees compliance with safety regulations in the construction industry (Unpublished doctoral dissertation). Purdue University, West Lafayette, IN.

Wilkins, J. R. (2011). Construction workers' perceptions of health and safety training programmes. Construction Management and Economics, 29(10), 1017-1026. doi:10.1080/01446193.2011.633538

Cite This Article as: Teachers College Record Volume 115 Number 1, 2013, p. -
https://www.tcrecord.org ID Number: 17002, Date Accessed: 12/8/2021 8:01:39 AM

Purchase Reprint Rights for this article or review
Article Tools
Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • James Wilkins
    Teachers College, Columbia University
    E-mail Author
    JAMES WILKINS is a workplace training consultant, instructional technology professional, and occupational health and safety researcher at Teachers College, Columbia University.
  • Marilyn Alibutod

    E-mail Author
    MARILYN ALIBUTOD is a bilingualism and bicultural education specialist based in New York City. She has worked with a diverse set of learners from a wide range of linguistic backgrounds and enjoys developing learning programs for second language learners. Her research program began with a pilot study that examined how identifying student orientations of integrative or instrumental motivation could be used to facilitate struggling students in their second language. The study was conducted in a dual-language (English-Spanish) classroom. She presents an extensive academic background in bilingual and bicultural education, elementary education and Spanish.
  • Anindita Nugroho
    Teachers College, Columbia University
    E-mail Author
    ANINDITA NUGROHO is a researcher in education policy and program evaluation. She has worked as a research consultant in five countries and is currently pursuing further graduate studies in economics and education at Teachers College, Columbia University. Her work is focused on quantitative methods and on international development.
Member Center
In Print
This Month's Issue