Development of a scale for recruitment of forest fire workers using confirmatory factor analysis

Forest fires cause serious damage to the ecological balance worldwide, threatening both property and the safety of society. Forest firefighting activities are defined as technical and administrative activities aimed at controlling, cooling and completely extinguishing fires, which tend to spread freely and can burn the whole or part of a forest ecosystem (Coşgun 2022). Forest firefighting activities involve complex decisions that take into account factors such as safety, cost, sociopolitical expectations, fire effects, concerns and ecological conditions (Thompson et al. 2018; Erdönmez et al. 2023). In other words, fighting forest fires is a complex, dangerous and honourable process that requires efficient organisation, collaboration among multiple teams and physical conditioning (IFSTA 2019; NWCG 2022; Heeren et al. 2023). In Turkey, forest engineers, forest rangers, technicians and fire workers are the primary workers in the fight against forest fires.

Forest fire workers (FFWs) play a crucial role in safeguarding natural resources, structures (Heeren et al. 2023) and various infrastructure elements, often at the risk of their own lives (Jolly et al. 2019). FFWs encounter numerous physical and psychological challenges in the course of their duties (Bos et al. 2004; Gordon and Larivière 2014; Marques-Quinteiro et al. 2022). During efforts to combat forest fires, FFWs frequently face work accidents and occupational diseases, including injuries, health problems and sudden premature death (Bos et al. 2004; Akay and Yenilmez 2007; Hauke et al. 2011; Kurlick 2012; Samsudin et al. 2021; Şafak et al. 2024). To prevent work accidents and occupational diseases, it is essential to manage the recruitment process correctly and carry out training and awareness activities effectively (ILO 1988; Hauke et al. 2011; Wang et al. 2019). The effectiveness of firefighting is significantly influenced by the training, experience, knowledge adequacy and adaptation to new technologies of the personnel engaged in forest fire control (Avcı and Korkmaz 2021; Şafak et al. 2023).

Personnel affect (ILO 1988) the quality of work and operational performance of a business. Therefore, understanding the personnel selection process correctly and implementing it with appropriate methods contributes greatly to an institution’s achievement of its strategic goals and expectations (Altun and Kovancı 2004). Personnel selection is a crucial process in which the best candidate is chosen using various techniques from a large number of candidates (Arsu and Uğuz Arsu 2021). Personnel selection is one of the critical decision-making problems in which managers consider numerous quantitative and qualitative factors (Chen and Hung 2020).

To meet the demand for a qualified workforce, the Vocational Qualifications Authority of Turkey (VQA) develops national occupational standards and qualifications, aligning with both the European and Turkish qualifications frameworks, outlining the minimum requirements for knowledge, skills and attitudes in various professions (Porsuk 2021). Despite the existence of professional standards for occupations such as forest production workers, non-wood forest product collectors, and forest cultivation and maintenance workers, the VQA has yet to create a national occupational standard and qualification document for FFWs in Turkey (VQA 2023). This not only hinders the establishment of clear standards for FFWs but also impedes the effective management of their employment processes in Turkey.

Several studies in Turkey have criticised the General Directorate of Forestry (GDF) for its inadequate management of the recruitment process for forest firefighters (FFWs). For example, Erdönmez et al. (2023) highlighted that some FFWs were unable to effectively combat large fires in 2021 owing to a lack of vocational training. Şafak et al. (2024) discovered instances where FFWs had a height of 155 cm and a weight of 45 kg. Additionally, only 37% of FFWs had body mass index values within the normal weight range, with 1% being underweight and 17% being obese. The authors explained these non-compliances as indications that the selection of FFWs is not being carried out properly. Kılıç (2012) criticised the physical unfitness of the majority of FFWs for firefighting duties, citing their overweight status and a perceived disconnection from the system. Both Kılıç (2012) and Ünver Okan and Acar (2017) noted that at least 70% of firefighting personnel had only received primary education. Additionally, Gümüş and Türk (2011) and Bacı and Çalışkan (2022) reported that 20.3 and 32.1% of workers, respectively, suffered from psychological disorders, deeming these data made them unsuitable as FFWs. Şafak et al. (2023) revealed that 27% of FFWs were assigned without any training on how to combat forest fires, emphasising insufficient training in forest firefighting activities and first aid. These deficiencies collectively underscore a systematic problem in the selection of FFWs in Turkey.

The qualifications required for the recruitment of FFWs should be established and defined in advance. Although the current literature addresses these criteria piecemeal, a consolidated overview of these criteria is presented in the following 15 items.

One of the most significant challenges individuals face in today’s world is undoubtedly stress and its associated factors. FFWs experience elevated levels of both physical and mental stress during their duties (Zafer 2016; Schmit and DeBeliso 2019). Some people have a risk-prone personality that leads to dangerous behaviour in the workplace. Hence, it is crucial to assess employees’ personalities at the initial stage of recruitment (Heydari et al. 2022).

Forest firefighting personnel come from heterogeneous backgrounds in terms of their education levels, origins (rural–urban life), family livelihoods (agriculture, tradesmen, wage earners, etc.), geographical region, etc. (Tadesse and Seboko 2013). For this reason, during the recruitment process, efforts should be made to align the knowledge level, experience, perceptions, preferences and attitudes of the personnel assigned to fight forest fires, fostering a more homogeneous composition. This approach will increase the effectiveness of equipment, decision support tools and training to be used in the future (Şafak et al. 2023).

GDF is responsible for all activities to be carried out within the scope of the organisation of fighting forest fires in Turkey (GRT 2018). GDF determined seven application criteria for fire workers to be hired as forest workers and firefighters in 2022 (GDF 2022a). These are: (i) forest workers must be between 18 and 25 years old, and firefighters must be at most 30 years old; (ii) applicants must be male; (iii) forest workers must have a minimum of primary school education and a maximum of secondary school education; (iv) firefighters must be graduates from a firefighting and fire safety department in secondary education, or must be a graduate of one of the associate degree programs in the field of firefighting, fire safety, civil defence, emergency, disaster management, or natural disasters; (v) a health report is required from the fields of ‘internal medicine, general surgery, ophthalmology, ear, nose and throat, neurology, and psychiatry’; (vi) applicants must be at least 167 cm tall; and (vii) there should be a weight difference of at most ±10 kg between the weight and the part of the neck that exceeds 100 cm (for example, if the applicant is 167 cm tall, his weight should be between 57 kg and 77 kg). The health report required for the employment of FFWs in Turkey does not contain a detailed evaluation in terms of suitability for their employment, which is considered one of the most dangerous jobs. This report only states that the candidate is suitable for routine work. Additionally, the candidate’s suitability for the job is not assessed based on the specific characteristics required for FFWs (physical, psychological and behavioural criteria, as well as conditions for training and self-management) outlined in the aforementioned 15 items.

In the forest ecosystems of the Mediterranean Basin including Turkey, an imminent increase in the danger and risk of forest fires to both people and assets is anticipated due to the impact of climate change. This situation will increase the occupational health and safety risks for FFWs (Jolly et al. 2019). Therefore, determining the criteria to use in the employment of FFWs, who are one of the stakeholders and involved in extinguishing forest fires, will help reduce the number and effects of work accidents and occupational diseases. The formulation of these criteria will contribute significantly to enhancing Turkey’s firefighting capacity.

Studies on personnel engaged in forest firefighting activities in Turkey generally focus on the following topics: satisfaction of FFWs with their workwear (Ünver Okan and Acar 2017), diseases, occupational health and safety of FFWs (Akay and Yenilmez 2007; Sayın et al. 2014; Bacı and Çalışkan 2022; Şafak et al. 2024), and the training of FFWs (Şafak et al. 2023). As evident from these studies, there is currently no research aimed at determining the criteria to be employed in the recruitment of FFWs.

This article aimed to develop a set of criteria that can be used in the recruitment of FFWs, which is an important problem in Turkey. The study adds value to the forest firefighting literature in two aspects. Firstly, it seeks to formulate a set of criteria designed for the recruitment of FFWs engaged in combatting forest fires. Secondly, it employs confirmatory factor analysis (CFA) to assess the validity and reliability of these criteria.

In this section, the study sample, sample size determination, survey form, criteria, explanatory factor analysis (EFA), reliability and CFA are explained within the scope of the research.

Descriptive statistical information on ground team workers participating in the survey is presented in Table 2. According to the data, 56.6% of the study participants were FFWs, while 43.4% were drivers or operators. Furthermore, 46% of the drivers and operators were fire truck drivers, and 40% were initial responder vehicle drivers. The youngest participant was 21, the most experienced was 72, and the average age was 42.4 years. Examining education level, it was found that 45% of participants have received at least a high school education. Additionally, 87% of participants were married.

The frequency, arithmetic mean and standard deviation values of the responses given using the nine-point scale to the 30 criteria determined for the recruitment scale are presented in Table 3. Accordingly, the average importance scores given to the criteria vary between 5.31 and 8.51.

CFA results

The path diagram shows the relationships between the variables of the CFA model, the factors and the effects of these factors on the observed variables (Kline 2005; O’Rourke and Hatcher 2013). In other words, the path diagram is a visual representation of the CFA model and is used to understand the theoretical structure of the model and evaluate the suitability of the model. The path diagram for the standardised results from the CFA is illustrated in Fig. 1. F_i (F₁–F₅) values shown as circles in the path diagram represent unobserved (latent) variables (factors). The ν_i (v₁–v₂₉) values shown in rectangular boxes represent the observed variables (criteria). The e_i (e₁–e₂₃) values shown as ellipses represent measurement errors. Two-ended oblique arrows connecting the rectangles indicate the observed correlations between factors in the standardised results. One-way arrows indicate the assumed influence and direction of one variable on the other. Statistical estimates of these effects are called factor loadings and are interpreted as regression coefficients, which can be in unstandardised or standardised form (Kline 2005). The two-headed oblique arrow between the error terms (e₂–e₃ and e₁₁–e₁₂) represents the covariance links (Maat et al. 2015).

Fig. 1.

Standardised CFA results.

The analysis results of the CFA model (model fit, estimates, and modification indicators) are examined to determine whether the EFA model is validated or not. As seen in Fig. 1, there is a weak same-directional relationship between the latent variables F₂, F₃ and F₄ (0.24–0.35). There is a strong (0.80) correlation in the same direction between F₁ and F₅. There is a moderate (0.58, 0.51) relationship in the same direction between F₄–F₅ and F₁–F₄, respectively. There is a same-directional relationship below the moderate level between the other latent variables (0.40–0.45). Additionally, when the covariance results and P-values (P = 0.001) for the latent variables were examined, no path was found to be removed from the model.

In Fig. 1, the regression weights between latent variables (F₁–F₅) and observed variables (v₁–v₂₉) are between 0.51 and 0.95, and P-values of all latent variables (F₁–F₅) are at the level of 0.001. For the observed variables to remain in the CFA model, the regression weights must be greater than 0.50, and the P value must be less than 0.05. Accordingly, there are no variables to be removed from the CFA model (Table 5 and Fig. 1).

Table 5.CFA results.

Structural relationships			β0	β1	s.e.	Critical ratio	P
v3	←	F1	0.721	1.000
v2	←	F1	0.734	0.947	0.050	18.870	0.001
v1	←	F1	0.723	0.849	0.046	18.566	0.001
v4	←	F1	0.842	0.982	0.045	21.625	0.001
v5	←	F1	0.880	1.058	0.047	22.630	0.001
v6	←	F1	0.836	0.996	0.047	21.268	0.001
v7	←	F1	0.756	0.987	0.051	19.236	0.001
v10	←	F2	0.874	1.000
v9	←	F2	0.889	0.968	0.032	30.330	0.001
v8	←	F2	0.754	0.798	0.035	23.001	0.001
v12	←	F2	0.582	0.764	0.047	16.206	0.001
v15	←	F2	0.509	0.616	0.045	13.679	0.001
v11	←	F2	0.592	0.524	0.032	16.609	0.001
v27	←	F3	0.659	1.000
v28	←	F3	0.801	0.805	0.055	14.651	0.001
v26	←	F3	0.688	0.658	0.047	13.871	0.001
v29	←	F3	0.530	0.728	0.062	11.788	0.001
v23	←	F4	0.793	1.000
v22	←	F4	0.955	1.014	0.057	17.674	0.001
v14	←	F5	0.743	1.000
v13	←	F5	0.755	0.884	0.046	19.238	0.001
v17	←	F5	0.696	0.842	0.050	17.002	0.001
v18	←	F5	0.769	0.948	0.050	18.924	0.001

β₀, standard regression values; β₁, non-standard regression values.

In the first CFA model, covariance matrices were used, and the fit indices that are frequently used in the literature, (χ²/d.f., RMSEA, CFI, and NFI) were examined (Erdoğan et al. 2007). In the pressent analysis, the χ²/d.f. value is 4.212. Additionally, the RMSEA value of the CFA model is 0.069, the NFI value is 0.896 and the CFI value is 0.919. The analysis shows that the fit indices are acceptable. However, modification indices were examined to reduce - χ²/d.f. and RMSEA values. Then, error covariance was added between e₁₁ and e₁₂ (60.800) in Factor 2, and between e₂ and e₃ (47.538) in Factor 1.

In the second CFA model, χ²/d.f. and RMSEA values decreased and reached acceptable values. In this analysis, the chi-square value (813.213) was found to be significant at the 0.05 level. The degree of freedom (d.f.) of the model is 218, χ²/d.f. value is 3.730, the RMSEA value is 0.063, the CFI value is 0.931 and the NFI value is 0.909. These results indicate that the fit indices are at a very good level.

The results described above indicate a high level of concordance between the model’s predictions and the actual data, supporting the assertion that the CFA model serves as an appropriate and accurate representation of the scale. Consequently, it was concluded that the CFA model applied to the FFW recruitment scale was valid.

This study aimed to determine a set of criteria for the recruitment of FFWs and to create a roadmap to address gaps in this context. A five-factor structure (suitability for work, physical and mental condition, education, working conditions and self-management) was established with goodness-of-fit indices at acceptable levels. These criteria can serve as foundational data for future studies aiming to develop basic performance indicators for FFWs.

The average importance scores given to the 30 recruitment criteria determined for FFWs, according to a nine-point scale, vary between 5.31 and 8.51. In other words, no criterion was generally considered unimportant or less important among these criteria. Accordingly, the most important recruitment criterion for FFWs is v₁ (Must have a suitable health status for work) with an importance score of 8.51. The second most important criteria are v₄ (Should be inclined towards teamwork) and v₆ (Should have a reputable and reliable character) with an importance score of 8.44. Third is v₅ (Must be business-oriented) with an importance score of 8.40. The three least important criteria are v₁₆ (Should be thrill-seeking and adventurous) with an importance score of 5.31, v₃₀ (Should have graduated from associate degree programs in the field of firefighting, etc.) with an importance score of 5.34, and v₁₂ (Must be competitive) with an importance score of 6.61.

An item (v₁₆ criterion) was added to test whether the items in the questionnaires were understood by FFWs and filled out reliably. The v₁₆ criterion, introduced for testing purposes, is not included among the criteria in the EFA model. This result shows that the questionnaires were filled out with understanding and appropriately by FFWs.

The scale developed in this study can be used for several purposes: establishing national occupational standards and national qualification documents for FFWs, as summarised by VQA (2023) and Porsuk (2021); selecting personnel for hand teams when contracting private companies for firefighting services, as noted by Avcı and Korkmaz (2021), and choosing permanent FFW personnel. In addition, in the selection of hand team personnel to be hired within the scope of service procurement from private companies, priority can be given to those who have volunteered in forest fires.

As mentioned by Hauke et al. (2011), it is impossible to entirely eliminate the risks faced by personnel engaged in fighting forest fires. Nevertheless, the implementation of an effective management system, including the selection and training of firefighting personnel, and compliance with occupational health and safety legislation, among other factors, can significantly enhance personnel protection, mitigate the intensity of work accidents and improve overall productivity. The scale developed and recommended for use as a result of this study has the potential to increase the GDF’s capacity and effectiveness in fighting forest fires.

Many fire departments conduct various physical, mental, safety and fitness exams to determine the motivation of firefighter candidates and to evaluate the competencies of firefighters (Heydari et al. (2022). In this context, the qualification exam practices outlined by Leduc et al. (2022) and NWCG (2022) for other countries should be reviewed by -GDF, taking into account the working conditions of FFWs in Turkey. Subsequently, a minimum set of criteria for employing both current and newly recruited FFWs should be established, taking into account the criteria outlined in the present study. Lastly, FFWs should undergo an annual evaluation for compliance with this set of criteria before the fire season.

Incorporating personality test results into the recruitment process can help ensure that individuals with high intrinsic values are selected as employees (Aydın Göktepe et al. 2020). Intrinsic value criteria corresponding to specific variables within the F₁ (Suitability for the job) and F₅ (Self-management) factors of this study serve this purpose.

Forest firefighting personnel are likely to encounter stress and similar situations (Zafer 2016). For this reason, it is necessary to assess the stress and danger resistance levels in FFWs before their recruitment (v₁₉). However, this v₁₉ criterion was excluded from the analysis owing to its failure to meet the conditions of EFA. However, criteria v₁₄ (Should have the ability to control anger), v₁₇ (Must be careful) and v₁₈ (Must be highly motivated) can be considered as equivalent to criterion v₁₉ to some extent.

FAT (2021) recommended choosing FFWs from graduates of associate degree programs that offer training in areas such as firefighting and civil defence. In line with this recommendation, criterion v₃₀ was added to the study. However, this criterion was later excluded from the analysis owing to its failure to meet the conditions of factor analysis. In this context, the majority of FFWs (54.7%) reported having received primary education, which may have introduced bias in reflecting their attitudes toward this criterion.

In Şafak et al. (2023), it is suggested that newly employed FFWs should be selected from those with basic knowledge about fighting forest fires. In a similar vein, Gümüş and Türk (2011) propose that employing trained, healthy, physically and mentally strong, and young individuals in the fight against forest fires can reduce the risk of workers’ injuries and work accidents, thereby increasing work efficiency. These two articles support the results of the current study.

Large forest fires have long-term negative and devastating social and economic impacts on ecosystems, habitats and communities (Tedim et al. 2018). For this reason, especially in efforts to fight large forest fires, the resources of ground teams (such as personnel and machinery) are just as crucial as those of air teams. Around the world, efforts to obtain support and benefit from volunteers in fighting natural disasters and forest fires have attracted attention in recent years (Martínez et al. 2021). Volunteers can be utilised effectively, especially in large forest fire situations, either as firefighters or support personnel. In this context, in the selection phase of volunteers to be assigned to forest fires, the recruitment scale developed in this study and the scales developed by Martínez et al. (2021) for the selection of volunteers to be assigned to natural disasters (until a new scale specific to volunteers in Turkey is developed) can be used. This volunteer selection system is essential to prevent occupational accidents that could affect the volunteers themselves or others, as highlighted by Whittaker et al. (2015).

In this research, a comprehensive literature research was conducted to determine the selection criteria to be used in the recruitment of FFWs to be assigned to fight forest fires. It was found that the current recruitment criteria used by GDF in Turkey for FFWs are insufficient. Therefore, by applying EFA and CFA procedures, a FFW recruitment scale consisting of five factors and 23 criteria was developed. The recruitment scale should be taken into account by VQA and GDF in the preparation of the national occupational standard and qualification for FFWs. GDF’s capacity to fight forest fires will increase with the establishment of a national occupational standard and qualification for FFWs. Additionally, adopting the new employment policy suggested in this study, which supports qualified workers, is expected to reduce the risk of work accidents and enhance the effectiveness of fighting forest fires. To increase personnel safety and firefighting efficiency in forest fires, research should be supported to establish a system for measuring, monitoring and evaluating personnel effectiveness in the coming years. Consequently, a multidimensional, reliable and valid recruitment scale has been designed, and it would be appropriate to use this scale in the recruitment of FFWs by GDF. Furthermore, the FFW recruitment scale can be utilised following initial evaluations in high forest fire risk countries (especially in the Mediterranean region). In this regard, the validity of the scale should first be assessed by considering cultural and language differences, followed by testing its reliability.