RiSK ASSeSSMent FOR HAzARd expOSuRe And itS cOnSequenceS On HOuSing cOnStRuctiOn SiteS in LAgOS, nigeRiA

Despite housing construction’s economic contribution, the nature of work done is well acknowledged as risky to execute because of the occupational accidents and work-related hazards to which workers are exposed. Most of the workers experience hazards, owing largely to inadequate or lack of safety infrastructure and mechanisms for protective gear. This article examines varying levels of hazards to which workers are exposed at housing construction sites in Lagos, Nigeria. A mixed methods research was used to collect the necessary data for the study. From the total number of 511 residential building construction sites identified, simple random sampling technique was used to select 255 (50%) of the buildings. A questionnaire was administered to the supervisors on each site to obtain information on the exposure of hazards on housing construction sites. In addition, one month’s data on incidents of near miss, accident and fatal cases were obtained from construction managers/supervisors for each site. The data was analysed with frequencies, percentages and inferential statistics. Construction workers are exposed to multifaceted hazards. Roughly 91% of the respondents had witnessed Acta Structilia 2020: 27(1) 60 hazards of varying degrees. Paired t-test values showed that, on average, 25.3 more near misses and 12.4 more accidents happened monthly on sites supervised by individuals/ owners than on sites supervised by trained supervisors. The Pearson’s r test (r = -0.705) showed that not enough first-aid kits were provided on sites to meet the needs of workers. Proper safety mechanisms to ensure strict adherence to safety rules and regulations at construction sites must be developed and enforced.

RiSK ASSeSSMent FOR HAzARd expOSuRe And itS cOnSequenceS On HOuSing cOnStRuctiOn SiteS in LAgOS, nigeRiA

Peer reviewed and revised March 2020
Published June 2020

*The authors declared no conflict of interest for the article or title
AbStRAct Despite housing construction's economic contribution, the nature of work done is well acknowledged as risky to execute because of the occupational accidents and work-related hazards to which workers are exposed. Most of the workers experience hazards, owing largely to inadequate or lack of safety infrastructure and mechanisms for protective gear. This article examines varying levels of hazards to which workers are exposed at housing construction sites in Lagos, Nigeria. A mixed methods research was used to collect the necessary data for the study. From the total number of 511 residential building construction sites identified, simple random sampling technique was used to select 255 (50%) of the buildings. A questionnaire was administered to the supervisors on each site to obtain information on the exposure of hazards on housing construction sites. In addition, one month's data on incidents of near miss, accident and fatal cases were obtained from construction managers/supervisors for each site. The data was analysed with frequencies, percentages and inferential statistics. Construction workers are exposed to multifaceted hazards. Roughly 91% of the respondents had witnessed hazards of varying degrees. Paired t-test values showed that, on average, 25.3 more near misses and 12.4 more accidents happened monthly on sites supervised by individuals/ owners than on sites supervised by trained supervisors. The Pearson's r test (r = -0.705) showed that not enough first-aid kits were provided on sites to meet the needs of workers. Proper safety mechanisms to ensure strict adherence to safety rules and regulations at construction sites must be developed and enforced.

intROductiOn
The housing construction industry is one of the largest employers globally and can contribute up to $10.5 trillion to the world economy by 2023 (MGI, 2017: online;NAPBHR, [n.d.]: online). It employs approximately 7% of the global work force (ILO, 2005: 6;MGI, 2017: online;Rhodes, 2019: 3-4) or 180 million people, and it is predicted to account for approximately 10% of the Global Domestic Product (GDP) by 2020 (Murie, 2007: 6-7;Durdyev & Ismail, 2012: 884-885;Nieuwenkamp, 2016: online). Owing to its substantial contribution to the GDP of the vast majority of nations, the development of the housing construction industry has been viewed as the foundation for contemporary and future economic growth and social development (Agbola, 2005: 7-10;Kasim, 2018: 955-959).
Despite its economic contribution, the nature of work done within residential building construction is well acknowledged as risky to execute because of the complexity of its activities and risky external and internal environments (Zou, Zhang & Wang, 2007: 602-603;Lette, Ambelu, Getahun & Mekonen, 2018: 58). The complexities of housing construction activities during execution expose workers to a plethora of hazards and generate enormous risks that may culminate into accidents if not well managed (Orji, Nwachukwu & Enebe, 2016: 282-283).
Absence of a centralised safety agency (Agwu, 2012: 213), inadequate or lack of safety infrastructure, regulations and mechanisms on sites, in the developing countries (Nigeria included), are responsible for the death of one person on site every five minutes (Takala et al., 2014: 326 Although a number of studies has been done on construction hazard and risk assessment in general (Babovic, 2009: 22-26;Oladinrin, Ogunsemi & Aje, 2012: 50-60;Zolfagharian Ressang, Irizarry, Nourbakhsh & Zin, 2011: 151-161;Orji et al., 2016: 282-289), there is a dearth of study on exposure to hazards specifically in the housing construction industry. The scarcity of scientific research on exposure to hazards in the housing construction industry highlights the need to assess the hazards and risks, taking into consideration the safety measures in place to reduce hazards. The purpose of this article is, therefore, to investigate the exposure to hazards and the consequences thereof on workers, working on housing construction sites in Lagos, Nigeria. Findings will be useful for all employers of construction workers, governmental and private employers, developers and other agencies that are involved in housing construction through the proffered safety measures towards mitigating hazard prevalence.

LiteRAtuRe ReVieW
To understand hazard exposure and its consequences on housing construction sites in Lagos, Nigeria, it is important to introduce the current theory on hazard exposure concepts included in this article. The existing theory focuses on the concepts of human error, hazards, and vulnerability of workers in housing construction.

Housing construction workers and human errors
The construction industry is an economically important industry in any country. The sector supply infrastructure and physical structures of a country provides basic needs such as housing for the population (Haupt & Harinarain, 2016: 80). The construction industry employs workers cutting across different occupations and trades. The categories of workers in typical housing construction include carpenters, masons, painters, electricians, machine operators (concrete mixer, crane, and soil compactor operators and forklift drivers), iron benders, plumbers, machinists (precision material worker who assemble or fabricate mechanical parts, pieces or products, using a variety of tools and equipment at construction sites), foremen and professionals such as architects, civil engineers, builders, foremen and urban planners. Jobs in the construction field require workers to hold various skills, from construction managers to floor installers. The skills are required as a gateway to the sector, but they can hardly protect the workers against hazards. The construction sector, including residential building construction sites, has a set of occupational hazards that are specific to the sector. Work done on construction sites is considered to be high risk and may result in occupational accidents that are mostly viewed as human-induced incidents or human error (Guo, Yiu & González, 2016: 5) Human factors engineering is a relatively new engineering field dedicated towards understanding human interaction in the work environment such as executing construction activities (Abdelhamid & Everett, 2000: 54). Human factor models capture worker interaction best in behaviour models that portray workers as being the main cause of accidents. This approach studies the tendency of human beings to make errors under various situations and environmental conditions (Abdelhamid & Everett, 2000: 54-55). In the construction industry, human error is always or sometimes attributed to accidents at construction sites, even at adherence or non-adherence to safety rules (Babovic, 2009: 22;Guo et al., 2016: 5-6). When the label 'human error' is used, it sometimes refers to the processes and systematic factors that influence people's behaviour and performance in any situation. Therefore, the scientific study of human error or failure should be concerned with the understanding of factors influencing the cognition, collaboration and behaviour of workers in a work environment (Woods & Cook, 2000).
However, it is important to holistically analyse the work environment with its inherent unsafe characteristics and analyse the design of workplaces and tasks that do not consider human limitations before the blame on workplace mishap can be labelled as 'human error' (Abdelhamid & Everett, 2000: 54-55). Human factors engineering aims to achieve efficiency and better designed tasks, using appropriate tools, in a safe workplace, but it also acknowledges limitations of human physical and psychological composition and capabilities which, if not properly managed, can trigger latent hazards and inherent vulnerability (Breeding, 2011: online).

construction hazards
A hazard is a situation that poses a level of potential threat or risk to life, health, property, or environment (CCOHS, 2020: online). The vast majority of hazards are dormant or potential, with only a theoretical risk of harm; however, once a hazard becomes active, owing to the relationship between a person and the work environment, it can lead to accidents and emergency situations (Breeding, 2011: online). Hazard exposure is driven by any activity, situation or condition within the physical environment with the potential to cause harm, injury or death to persons, and damage to assets and the environment (Rausand, 2004). Cardona (2001) notes that hazard refers to a latent danger or vulnerability to an external risk factor from an exposed system or subject.
According to OSHA (2017), health hazards in the construction industry can be grouped under chemical, physical, and ergonomic hazards. Chemical hazards can affect the body via inhalation, ingestion, or skin absorption, including toxic gas from welding; fumes from dusts; burns from chemicals; skin irritation from cement and paints, as well as respiratory irritation from thinners and insulation materials (OSHA, 2017: 3-7). Physical hazards are noise, heat, vibration and radiation, including acute noise generation; acute vibration; electrocution; exposure to excessive heat, and exposure to hot or pressurised liquid (OSHA, 2017: 10-15). Ergonomic hazards include mainly manual handling of loads such as objects falling from a height; fall of workers from a scaffold; fall of workers from a roof or upper floor of a building; workers pierced by sharp objects; workers falling into unfenced excavation, and workers struck by earth-moving machines (OSHA, 2017: 27-29).
In the context of this article, a hazard is defined as any form of mishap or injury suffered by workers as related to specific occupational demands or job requirements on site.

construction hazards management
Scope, environment and working conditions on construction sites can trigger inherent hazards that could be managed, using safety management approaches. Inadequacies in the safety system could influence and/ or lead to occupational accidents (Ford & Tetrick, 2011: 49-50;Andersson, 2012: 210-212). For example, poor work ethics and flagrant disobedience to safety rules and regulations by workers would lead to accidents, regardless of the existing safety management approach. Human-and, to a larger extent, non-human-related events are beyond control and prediction, as encapsulated in the safety management concept. Hence, as observed by Imriyas, Phen and Teo (2006: 272-274), estimation of exposure to hazards and the functionality of any occupational injury risks analysis in construction projects should be assessed, using two factors, namely a project's safety management level and inherent hazard level (Imriyas et al., 2006). Figure 1 shows that, in construction safety management, inherent hazard level identification is fundamental, because unidentified hazards are the most unmanageable risks (Carter & Smith, 2006: 198). As shown in Figure 1, unmanageable and unidentified hazards would gravitate a construction project towards the accident-and hazard-related zone, while informed safety measures would pull the construction activities towards the safety zone. In addition, when and where the safety management force is equal to the degree of hazard generated, the construction activities would be inclined towards the neutral zone. However, if the safety management structure is lower than the hazard level, the project would be driven towards the accident zone. Hence, the linkage and/or the assessment of a construction project hazard level and safety management structure prediction of occupational accidents inherent within the environment of a project.
Bosher and Dainty (2011: 1-3) observe that inherent hazards could lead to risks and these identified risks could be minimised, transferred, shared, accepted or managed, but should never be ignored. Fundamentally, two phases are required in the control and management of construction hazards. First, the hazardous events prevention phase and, secondly, mitigating and adapting to the potential severity of hazards, if and when they occur (World Bank, 2013: 6-10). Varying degrees of hazards occur at construction sites. Their degree of severity or fatality is dependent on the safety measures taken by workers on site. At construction sites, preventive measures are taken for different probable types of hazards such as protective boots against piercing; protective helmet against fall impact on the head; nose mask to filter fumes and dust, and to protect against respiratory diseases, and so on (ILO, 1999: 79-83).

ReSeARcH MetHOdS
This study assessed the hazards faced by workers on housing construction sites in Eti-Osa, Lagos State, Nigeria, taking into consideration the level of exposure to, and the impact of the identified construction hazards. The study used a mixed methods design, in which qualitative and quantitative data are collected in parallel, analysed separately, and then merged (Creswell, 2014). A quantitative structured questionnaire survey enables researchers to generalise their findings from a sample of a population (Bryman, 2012: 232;Creswell, 2014). It also allows for descriptive and inferential statistical analysis (Naoum, 2013: 104). In this study, hazard exposure records of housing construction sites were used to build the theory on hazard exposure, predicting that experience of 'high-risk' hazards will negatively affect the health of workers on housing construction sites in Nigeria. The questionnaire assessed the level of hazards experienced by workers. The reason for collecting both quantitative and qualitative data is to elaborate on specific findings from the breakdown of the hazards incident records, such as similar 'high-risk' hazards experienced from the respondents' groups (Creswell & Plano-Clark, 2007).

Sampling method and response rate
From a preliminary survey by the authors, a total of 511 new residential building projects, at different stages of construction, were identified in the study area. Random sampling was utilised to select 255 of the residential buildings under construction. From these, at least one worker acting as supervisor/construction manager or owners' representative from each site was targeted to participate (Alvi, 2016: 35). The survey was undertaken with a sample of 255. The sample size for research done in construction-related populations was calculated in accordance with the table recommended by Krejcie and Morgan (1970: 608). From the table, the recommended sample size for a population of 500 is 217, and for 550, 226. This recommendation validates the sample size of 255 as efficient for the population of 511 in Table 1.

data collection
Structured questionnaires were used to collect data from 255 construction managers or owners' representatives (supervisors) working on new residential building construction sites in Eti-Osa, Lagos, Nigeria, from 2 to 17 November 2016.
The questionnaire consisted of five sections. The first section, on the respondents' demographic profile, obtained personal information on age, educational level, and occupation. The second section set 14 tick-box options on hazard experiences to obtain if and how often workers experience hazard situations on site. Section three set one Likert-scale question with 18 items on the construct 'exposure to hazard' and section four set one Likert-scale question with 18 items on the construct 'consequence of hazard exposure'. Participants were requested to rate the level of frequency and the level of consequence on the statements regarding hazard exposure on housing construction sites. Section five contained four tick-box questions and one Likert-scale question with 5 items on the construct 'control strategies'. Participants were requested to rate their level of agreement on statements regarding safety management of hazards on site. The results from these measurements form the items used in the descriptive analysis, matrix analysis, and the inferential statistics. To reduce the respondents' bias, closed-ended questions were preferred for sections two and three (Akintoye & Main, 2007: 601).
In addition to the questionnaire survey, supervisors also provided a tabulated record sheet of one month's data (2 November to 1 December 2016) on incidents of near misses, accidents and fatalities in all the sampled sites. All the 255 copies of the questionnaire administered were returned and used for analysis. The authors also recorded their own observations on safety management from the sites sampled.

Analysis method and how to interpret data
The Statistical Package for Social Science (SPSS) version 24 was used to determine the frequency and consequences of hazard exposure, by using descriptive and inferential statistics (Pallant, 2013). The frequencies and/or percentages of responses were generated and reported, in order to analyse the respondents' profile, supervisor status and exposure to hazards, incidents records and safety management structure.
For the analysis on the frequency, and consequences of hazard exposure, a 5-point Likert scale was used to measure how strongly respondents felt regarding the statements in the Likert-scale constructs. Likert-scale rankings are effective where numbers can be used to quantify the results of measuring behaviours, attitudes, preferences, and even perceptions (Wegner, 2012: 11;Naoum, 2013: 89). For the purposes of analysis, it is important to note that the following scale measurement was used regarding mean scores (MSs), where 1 = very low/insignificant; 2 = low/minor; 3 = regular/moderate; 4 = high/major, and 5 = very high/catastrophic. Data was analysed, using frequencies and MS rankings.
To identify the major risks from hazard exposure, a risk matrix analysis was done to rank the 25 initial item scores on the basis of frequency and consequence. The matrix assessment is the multiplication of hazard exposure and its consequence: (hazard exposure x consequence). Only one score for each item (as labelled) was considered to determine the influence of each item in the row on the statement in the column. For example, both consequences and occurrences range from 1 to 5. The risk matrix score for a specific hazard is the product of the rate of occurrence (row value) and consequence (column value). Items in the matrix were weighted (scored) on a 5-point Likert scale were 1 = very low risk; 2 = low risk; 2 = moderate risk; 3 = high risk; 4 = high risk, and 5 = very high risk. They were then classified according to their risk level where 1-4 = 'low'; 5-15 = 'medium', and 16-25 = 'high', stipulated in the risk matrix assessment model of Zolfagharian (Zolfagharian, Nourbakhsh, Irizarry, Ressang & Gheisari, 2012: 1753) (see Table 2).
To analyse the hazards incidence records, a paired samples t-test with p=0.05 was done to compare the mean of cases, and to find any correlations between the cases examined (Ross & Willson, 2017: 17). A Pearson's correlation test with p=0.05 was done to find any significant relationship between workers' need for first-aid kits, and the provision of such kits by housing construction site managers.

Limitations
Owing to time constraint, focus-group discussions could not be organised in order to identify the sociocultural dimension influencing hazards in construction sites. A section of Lagos and construction activity was also sampled; this may not be a true reflection of construction activities in Nigeria. Table 3 displays the demographic profile of the participants. It is obvious that the majority of the respondents were foremen (19.4%), masons (16.9), carpenters (14.1%), and iron benders (11.4%); the professionals (architects, civil engineers, builders, and urban planners) constituted only 9.4% of the respondents. The results show that the respondents represented all relevant workers in the housing construction.

Respondents' demographics
The highest number (59%) of the respondents had a secondary school education (36%) or a primary school education (23%), and 12% had a technical college qualification; only 8% had a tertiary qualification. Furthermore, the analysis of the respondents' academic qualifications showed that 20% of the respondents did not have any formal education, especially the carpenters and the painters; they were trained through the artisanship system, the predominant informal training arrangement for tradesmen and artisans in Nigeria (Sanni & Alabi, 2008: 17;Adewale, Siyanbola & Siyanbola, 2014: 37-38). In addition, 19% of the foremen did not receive any formal education. Analysis of the respondents' age showed that 55% were younger than 40 years, except for the machinists, foremen, and machine operators, the majority of whom were above 40 years of age. From the information on the academic qualifications of the respondents, it can be inferred that the majority of the supervisors possessed satisfactory education qualifications to understand the questionnaire and supply data for this study.  Table 4 shows the status of supervisors as well as if and how often they experienced hazard situations on site. A vast majority (78%) of the residential building construction activities were supervised by owners (private individuals) or appointed cronies of the owners. As observed, most of these persons (74.3%) do not have prerequisite qualifications to supervise and manage housing construction projects.

Supervisor hazard experience
Registered construction companies were involved in the construction of 22% of the residential building projects, and almost 98.2% employed qualified personnel to supervise work on site. Construction activities are laden with hazards of varying magnitude and the level of exposure varies among the workers. Analysis of the respondents' experience with hazards showed that the majority of them (91.1%) had witnessed one form of hazard or other, while 9.9% had not witnessed any construction work-related hazard. From the total number of those who had witnessed hazards previously, 68.4% and 31.6% were supervised by persons without prerequisite qualifications in individual and corporate construction sites, respectively. A vast majority of respondents experienced hazards occasionally (45.5%) and frequently (47.2%).  Table 5 shows the ranking of the level of exposure to hazards as well as the consequences of the exposure, as perceived by the respondents. The majority of the respondents (53.7%) perceived very high exposure to objects falling from a height, and 49.3% indicated that falling objects from a height (MS = 3.99) constituted a major health risk to workers.

Ranking of exposure to hazards and its consequences
With an average MS of 3.24, results in Table 5 show that respondents perceived regular exposure to hazards on construction sites. With MS ratings above 4.0, respondents perceived high exposure to 'objects falling from height' (MS = 4.25) and 'skin irritation from cement, and paint' (MS = 4.17). The majority of the respondents (85.6%) indicated that 'skin irritation from exposure to cement and paint' (MS = 4.14) is a major and sometimes catastrophic health risk to workers. Almost three quarters of the respondents (73.1%) indicated high to very high levels of exposure to noise pollution. With MS ratings above 3.5, respondents are regularly exposed to 'piercing by objects' (MS = 3.79); 'air pollution' (MS = 3.67), and 'falling from a roof' (MS = 3.59). More than half of the respondents (60%) also indicated that 'falling from a roof' (MS = 3.62) and 'piercing by sharp objects' (MS = 3.57) may pose a major health risk to workers. Owing to the specialised nature of work to be done in electrical wiring and installation of electrical equipment in buildings, in addition to the fact that most of the work was done without active connections to electricity gridlines, 46.5% of the respondents considered exposure to electrocution to be very low and 43.8% of the respondents adjudged the consequence to be insignificant. Exposure to 'fall into trench' (MS = 1.85) and 'struck by a machine' (MS = 1.96) were rated very low, showing that at housing construction sites the consequences of exposure to these two hazards (MS = 1.84; 1.94) is insignificant.

Risk matrix results
Adopting the computation in Table 2, the observed risk matrix assessment shows the classification level of hazard exposure and its consequences in Table 6. With classification levels from 16 to 20, 'fall of object from a height', 'fall of workers from a roof or upper floor of a building', 'acute noise generation', 'exposure to excessive heat', and 'skin irritation from cement and paint' were classified as 'high-risk' hazards. Experience of these highrisk hazards will have a major effect on the workers' health.  Table 7 shows the records for near misses, accidents and fatal incidents recorded for the sampled housing construction sites. Supervisors recorded 1,213 hazard-related incidents for one month, showing that workers are exposed to 40 hazard experiences daily.

Hazard incidents results
Incident records also indicated that 631 near misses had been recorded (176 cases from private and 455 cases from corporate sites). Approximately 380 actual accidents had happened (257 cases from private and 120 cases from corporate sites); 205 fatal cases were recorded (122 cases from private and 83 cases from corporate sites). Falling objects from a height (249); piercing by sharp object (186); workers falling from scaffolds (175) and/or from roof or upper floor (155); exposure to excessive heat (143), and burns from chemicals (112) were hazards that occurred the most.    Table 9 shows the results on how safety control and management prevent hazards on sites sampled. Overall, only 7.8% of the supervisors provided regular medical orientation about the hazards to which construction workers were exposed in the course of carrying out day-to-day activities. The vast majority of the individual/owner supervisors (97.9%) did not provide any medical orientation drills to keep workers abreast of hazards associated with construction activities. Corporate supervisors (80.4%) and individual/ owner supervisors (only 5.1%) provided safety officers on site to monitor compliance with safety regulations in the sampled housing construction sites.

Safety management to mitigate hazards
Overall, more than half (64.7%) of the supervisors did not provide adequate emergency response plans and procedures to workers on site and 72.1% (the majority from the private/owner) of the supervisors did not provide the availability of on-the-job safety requirements and training to workers. Overall, only 5% of the supervisors indicated that they provided insurance cover for workers on sites. None of the workers from the private construction sites had insurance cover, while roughly 23.2% of the workers in corporate construction sites had insurance cover. In the sampled sites, only 27.4% had functional first-aid kits to respond to emergencies. In Table 10, the results from the Pearson's test show that there was a significant relationship between the number of first-aid kits required by construction workers and the number of first-aid kits provided by construction site supervisors [r = -.705*, n = 255, p = .005]. This implies that the number of first-aid kits needed on site to respond to worker emergency situations and the number of first-aid kits provided by housing construction site supervisors are not enough. The situation was extremely dire in the private/ownersupervised sites, where 87.9% of the sites did not provide first-aid kits for the workers. Site observations, during the course of the study, showed that in most sites visited, especially in owner-/individual-supervised construction sites, workers were not kitted up with the required wear and assemblage necessary to perform tasks assigned and to ensure personal protection from hazards. In addition, only a few made it mandatory for visitors to wear personal protective equipment before accessing construction sites.

diScuSSiOn
Housing construction activities in Lagos, Nigeria, are initiated and executed by private individuals, corporate organisations and government as part of infrastructural facility and social support to the citizens. As shown in this study, safety measures expected to be enforced on site is a function of the executor and quality of supervision. It is clear from the study that most of the supervisors did not possess the required qualification to supervise construction projects. This has implications on the capacity to identify risks, reduce hazards and enforce hazard-abatement measures, in order to avert accidents.
Findings revealed that 91% of the respondents had direct or indirect exposure to hazards of varying degrees. There are hazards that presume high risks, due to high or very high level of exposure (occurrence) and a major or catastrophic consequence. In this study, exposure and consequence of object falling from a height; workers falling from roof top or upper floor; noise pollution; exposure to excessive heat and skin irritation as a result of contact with cement and paint were classified as 'high-risk' hazards. The hazards that pose 'medium risks', owing to average exposure and moderate consequences, include fall of workers from scaffold; piercing by sharp object, and poor ambient air quality with toxic atmospheric condition from fumes and dust particles. 'Low-risk' hazards identified with rare or very low exposure with insignificant or minor consequences were electrocution; fall into unfenced excavation; struck by earth-moving machine; toxic gas from welding; burn from chemicals, and uncontrolled release of pressurised liquid.
The leading causes of private sector worker fatality, as noted by the United States Department of Labor (2017: online), in the construction industry included falls, struck by object, electrocution, and caught-in/between (compressed by equipment or objects, and struck, caught, or crushed in collapsing structure, equipment, or material). The four causes were collectively responsible for more than half (58.6%) of construction worker deaths. Except for electrocution and caught-in/between, the hazards with high chance of occurrence and with major consequences identified in this study fall within the leading causes of fatality. In addition, electrocution may have a very low exposure as shown in this study, but its consequence is always fatal.
As shown in the study, more than any other workplace activity, work at a height is very risky and accounts for more injuries and deaths yearly on construction sites (Nadhim et al., 2016: 638-640;Hamalainen, Takala & Kiat, 2017: 11-13;Vanguard Newspaper, 2017: online). Falls from roofs is one of the leading causes of construction work-related fatalities (Toscano, 1997: 38-40;Nadhim et al., 2016: 640; United States Department of Labor, 2017: online). It accounts for roughly one-fifth of all housing construction fatalities (Bosher & Dainty, 2011: 2-3;Ede, 2011: 156). Fall accidents in construction projects, particularly building works, are the most frequent accidents (Parsons & Pizatella, 1985;Gillen et al., 1997: 650-651;Latief, Suraji, Nugroho & Arifuddin, 2011: 81-81;Nadhim et al., 2016: 638-640). Housing construction workers must adhere to stage hierarchy to all work at a height. These are the avoidance of work at height, the prevention of personnel from falling, and mitigation of effect of falls, should falls occur. For example, assembling of components should be done at ground level. However, if an activity must be done at a height, site managers must take appropriate measures to prevent workers from falling from a distance that could cause injury and fatality. The use of guard rails and scaffolding platforms to prevent falls and individual measures such as safety harnesses must be enforced.
A plethora of hazards is associated with scaffold use and misuse (Davies & Tomasin, 1996). These include slippery conditions on the platform; defects in the members of the scaffold/ladder; overloading of materials and workers on the platform, and the nature of the platform on which the scaffold/ladder rests. However, observation from the sampled sites showed that most of the scaffolds, especially in the private-owned housing construction sites, were makeshift and the majority of these were constructed/made from bamboo. The design factors, particularly adequacy of design, have been compromised. This increases vulnerability to hazards, regardless of the safety management structure in place.
Overall safety management structures on housing construction sites sampled do not adhere to environment and occupational health and safety regulations of Nigeria. For example, 72.5% of the sites did not have firstaid kits available for workers needed to respond to emergency situations. In the private-supervised sites, 87.9% of the sites did not provide firstaid kits for workers and only 0.5% had safety officers on site to monitor compliance with safety regulations. This non-compliance by the private construction supervisors may be as a result of additional cost to be incurred in engaging such safety professionals. Housing construction activities is a capital-intensive venture, where the proponents usually seek avenues to reduce cost (Agbola, 2005: 16;Turok, 2016: 235;Alabi et al., 2018: 47;Kasim, 2018: 958). However, cost reduction should not override the dignity in labour and value of human well-being and for human lives.
The provision of safe, hazard-proof construction sites is a function of the level of planning and decisions taken by qualified personnel appointed to supervise work on site. Supervisors are technically excellent at their job, owing to years of experience, coupled with the engagement of workers who are also highly skilled, in terms of getting something built correctly. However, this level of competence is not a guarantee for safety and hazard abatement at the construction site. This is made more precarious when the majority of the supervisors are not qualified and may have the general notion, as shown in this study, that the construction site is unsafe and the hazards and risks, to which the workers are exposed, are the usual dictates of the work environment (ILO, 2014: 2; Guo et al., 2016: 5-6;Lette et al., 2018: 57). Therefore, nearly all construction hazards leading to accidents occur as a result of negligence on the part of the supervisor to impose supervisorworker responsibilities. The supervisors' laxity transfers hazard to workers on site, because the supervisor is charged with the responsibility of overall site safety. The deficiency in the enforcement of construction sites' health, safety and environment provisions manifests in workers carrying out responsibilities in blatant disregard for safety measures, by using inadequate and unsafe equipment in unsafe site conditions, as shown in this study.

cOncLuSiOn And RecOMMendAtiOnS
Workers are prone to different construction work-related hazards in Lagos, Nigeria. The hazards have been identified and, with the probability of occurrences and consequences of impact, classified as high-, medium-and low-risk hazards. To correctly avoid fatalities caused by these risk hazards, proper safety mechanisms are needed. In this study, these mechanisms include qualified professionals as site supervisors, the enforcement of safety regulations on the sites, and the usage of appropriate personal protective equipment for specific jobs.
Supervisors have the highest influence on site safety, because they coordinate, direct and monitor the work of all the subsectors within the construction sites. It should be mandatory for housing construction projects to be supervised by professionals with the requisite knowledge of construction hazards and safety measures. The enforcement of compliance with specific safety regulations and rules for different construction activities in the context of construction methods, materials and execution is very important.
Adequate and continuous training for construction workers through workshops, regular site meetings focusing on safety must be adopted. Laws regarding employers' responsibility towards employees for every site construction activity must be enacted and enforced. It is expected that workers and visitors alike in the housing construction sites should undergo a safety drill and be supplied with head-protection gear and mandatory head-protection signs and footwear displayed around the site, especially at private construction sites. For the sustainability of the contribution that housing construction projects make to the overall economy of Nigeria, a centralised regulatory framework to ensure that safety rules are acquired for every construction activity should be implemented.