Instrument for Measuring Intentions to Leave

Abstract

The aim of this study is to show the design of an instrument that contextualises the intention to leave (IL) in terms of spaces and job content (SJC), the perceived supervisor support (PSS), the perceived peer support (PPS), the organisational extrinsic motivation (OEM), the available opportunities outside the organisation (AO) and family–work conflict (FWC) as a tool to make diagnostic evaluations of Mexican hotel companies. The method used for the design of this instrument included literature review, establishment of dimensions and items and items validation. The reliability of the instrument was evaluated through the technique of split halves, and eventually it was calculated using Cronbach’s alpha coefficient in the different components of the instrument, as an indicator of internal consistency. The correlation coefficients obtained were acceptable in 38 of the 40 and significant at 0.01 level with only two at the 0.05 level. The tool proved to be reliable as a whole and in all dimensions in the first scan. Thus, the results revealed that this instrument measures the following: the causal dimensions, intention to leave and its variables.

Keywords

Intention to leave personnel turnover turnover intention perceived organisational support job content motivational tools voluntary turnover

Introduction

One of the main challenges facing the hotel industry is voluntary turnover. In tourist destinations, the hotel industry is extensively documenting a rotation that can range between 58 and 112 per cent annually (Denvir & McMahon, 1992; Pizam & Thornburg, 2000), and this becomes twice the average rate observed in other sectors (Deloitte, 2010). The causes attributed to the phenomenon vary, but the most common are the opportunity to access ‘a better job’, ‘family problems’, ‘problems with the supervisor’, ‘problems with peers’, ‘best performance’, among others (Blomme, Rheede & Tromp, 2010a; Aguirre et al. 2015; Eddleston, 2009).

Contextualising, the hotel industry has to consider that they are offering a personalised service based on employee services, because customers perceive quality and obtain satisfaction, through the contact they have with workers. Currently, high levels of stress, given the nature of the hotel industry for managing unpredictable schedules and work more than eight hours (Gamor, Amissah & Boakye, 2014), can foster conflict between family and work (Gamor et al., 2014) and can perceive lack of support from the supervisor (Karatepe, 2012) and even a mismatch between wage benefits, recognition perceived and the given work (Mohsin, Lengler & Aguzzoli, 2015). This may encourage the employee to seek better employment opportunities. The financial loss associated with the cost of employee’s turnover can reach 200 per cent of the total employee remuneration depending on the level of the position exercised by the employee (Deloitte, 2010), without considering the possible loss due to inconsistencies in the service offered. Therefore, keeping control on voluntary turnover becomes imperative (Dusek, Ruppel, Yurova & Clarke, 2014).

Therefore, this study establishes the main objective to design a scale or an instrument that contextualises the intention to leave (IL) in terms of their possible causes and that allows for a diagnostic evaluation in the field of Mexican hotel companies. In addition, it could allow one to contextualise the causes of employee turnover and intention for leaving. The instrument would allow one to understand the scope of the IL by articulating the conceptual and methodological components applied to a specific sector.

The term ‘scale’ is used to refer to a certain type of a measuring instrument, which has been used extensively in the field of social sciences (Kerlinger & Lee, 2002). As with other types of measuring instruments, the basic elements that form a scale are called items, and in this case they are typical situations that are presented to people in order to provoke a response-related attribute evaluated by the instrument.

Attitude scales are a measurement tool that allows us to approach the evaluative variability of individuals with regard to any object. In principle, it appears as a relatively simple tool. A set of responses are used as indicators of an underlying variable (attitude). In this sense, it is necessary to ensure that the properties of the indicator used correspond with that which we can assume or postulate, which belong to the variable.

The IL or labour turnover intention has been understood as the intention that employees have to leave an organisation to which they belong, leave or resign, despite having the opportunity to stay in it (Regts & Molleman, 2012). In addition, several studies have shown that staff turnover affects a variety of processes and organisational outcomes among which is the job satisfaction of the people, the work environment, performance and replacement costs (Li & Jones, 2012; Regts & Molleman, 2012; Zimmerman & Darnold, 2009).

In this context, this work is an effort to contribute to the study of the phenomenon of staff turnover from two perspectives: (a) The conceptual perspective of linking various factors on the behaviour of people at work with the turnover intention because both have been widely related aspects in organisational literature, although it has been empirically recognised that required integral approaches, and reference frames are used consistently with the nature of the observed phenomenon and (b) the methodological perspective that seeks to validate the instrument to measure the typical organisational factors that guide the intentions of people to leave in organisations, in an instrument of diagnosis. Recognising the limitations of this tool (variables used, reliability and validity) and the need for more research in different contexts and samples, for this stage, the two perspectives involve the main objectives of the study.

Furthermore, when considering the personnel turnover as a variable dependent on another, it can be seen that its results incorporate effects arising from a set of associated variables. In the first instance, we consider the IL or intention to resign, and then from this intention, arises voluntary turnover. The relationship between intention and action to leave has been documented in several studies (Griffeth, Hom & Gaertner, 2000).

The above-mentioned information points out the need to contextualise the theoretical and methodological designs used in generalities, but consider, especially, the particularities of the sectors and groups involved in the studies. The results of this type of exercise would identify critical or relevant elements for specific organisations or cultures and also enable the construction of scenarios and strategies for validation, generalisation and application of knowledge.

Conceptual Framework

Researches on the labour IL are often used to measure attitudes using Likert scales, in which the answers express a position on a posed question (Liu et al., 2013). Some handle the IL with polarities, such as, intention or not to stay (Cheng & Waldenberger, 2013), which ultimately would value the intent of a person to keep his job, as opposed to leaving it. In this sense, some studies have found up to 89 different variables of factors that contribute to or mediate with the IL (Cheng & Waldenberger, 2013).

On the theoretical level and in practice, there is no agreement on ways to refer and deal with the phenomenon. Some researchers (Adams & Beehr, 1998; Münderlein, Ybema & Koster, 2013; Schmidt & Lee, 2008) used interchangeably personnel turnover and retirement and even, in practice, are also used indiscriminately for dismissal or organisational strategy and voluntary departure. However, both the description of phenomena and the definition of methodological options for studies are important to note the differences, since, on the one hand, we can describe the phenomenon wrongly and, on the other hand, could be used to design inadequate instruments.

These requirements clearly are supported by some authors who argue that different forms of departure of an organisation also have different explanations, as per the age of employees, their present motivations, their expectations towards the future and the work and their interests and job outlook (Adams & Beehr, 1998; Schmidt & Lee, 2008). The importance assigned to each aspect is expressed also in the processes and results of their investigations.

Over time, the scientific and practical interest in personnel turnover has incorporated various dimensions and related perspectives. For example, it has been sufficiently recognised that the decision made by employees to leave a company carries high costs to the organisation (Wright & Bonett, 2007), on replacing staff due to the loss of knowledge. The experience and the lessons learned tell them that they have the human capital but totally unlinked. In addition, it has found that voluntary turnover can impair consolidation processes and teamwork (Siebert & Zubanov, 2009).

Although since the early 1980s, staff performance has become a focal point for the personnel turnover research (Allen & Griffeth, 1999; Bycio, Hackett & Alvares, 1990; Jackofsky, 1984; Steers & Mowday, 1981; Zimmerman & Darnold, 2009), and despite many efforts to understand the diversity and complexity of the causes of personnel turnover, important aspects of their relationships in organisations are still unknown. In an effort to dispel doubts in the light of the antecedents of employee turnover, researchers have focused on modifying risk factors with the expectation that changes in these factors affect in the same way, the intentions to leave and rates of personnel turnover thereof.

Furthermore, research consistently shows that the IL is the best predictor of personnel turnover (Griffeth et al., 2000; Van Breukelen, Van der Vlist & Steensma, 2004). In addition, the relationship between intention and effective action suggests the relevance of some conceptual and methodological tools that explain, in context, the phenomenon of staff turnover and unveil multiple interactions with different variables.

The literature reviewed shows that the IL organisations have multiple factors, as shown in 48 studies from China, the USA, Holland, Germany, Canada, India, New Zealand, Norway, Israel, Korea, Turkey, Australia, United Arabians Emirates, France and Zimbabwe. The search for relationships between organisational factors of different companies with different sample sizes and different types of jobs has formed a major or dominant approach where staff turnover is approached through IL organisations (Aguirre et al. 2015).

The multiple factors discussed in the study can be organised into categories, such as, work attitudes, attitudes towards the organisation and attitudes towards foreign and variables that have to do with the individual. The most influential factors in these categories are: space and content of work, perceived supervisory support, perceived peers support, available opportunities outside the organisation, work–family conflicts and organisational extrinsic motivators (Aguirre et al. 2015). A brief description of the factors allows one to visualise their possible interactions.

Spaces and Job Content

Workers have an affective perception of intrinsic and extrinsic job conditions that may include satisfaction with the spaces and aspect conditions of where they work. Aspects of job content and the way that is structured, and satisfaction with division of work (Münderlein et al., 2013). In this line, some studies have investigated about employee satisfaction through aspects of work in organisation, such as, the timing or distribution of tasks (Mcnall, Masuda & Nicklin, 2010), compensation (Lai & Kapstad, 2009), the physical conditions of work and its relation to organisational results as output intent (Blomme et al., 2010a).

Perceived Supervisory Support

Supervisors are typically the organisational agents closest to the employee and have the ability to communicate the intentions of the organisation directly to their subordinates. Therefore, it is stated that the support of the supervisor plays an important role in facilitating the employee job fit.

The perceived supervisor support (PSS) involves behaviours through which employees learn whether they are fit or not within the organisation. The supervisor is an agent of the organisation who helps embody the culture and can generate a perfect fit between labour and employee skills.

Workers can expect that supervisors have the principal responsibility to provide appropriate opportunities for learning and growth to increase their skills (Lai & Kapstad, 2009). Previous research revealed positive relationships among PSS and performance (Staufenbiel & König, 2010), promotion opportunities (Stahl et al., 2009), job satisfaction (Kyei-Poku & Miller, 2013), affective commitment (Chinomona & Chinomona, 2013) and job content (Münderlein et al., 2013) and, of course, with the IL the organisation.

Supervisors also have the ability to shape the perception of employees about their authority and responsibility. Finally, supervisors administer a microculture that influences the extent that employees establish contacts with peers. The employee perceptions about culture, management skills, peers, authority and responsibility are synthesised in the concept of ‘job fit’.

Perceived Peer Support

This aspect has been studied by some authors who show that the interaction with peers and closest colleagues plays a crucial role in explaining the motivation to learn new skills and position at work, seeking growth and stability in it (Biron & Boon, 2013; Stahl et al., 2009; Yang, Gong & Huo, 2011). The opportunities to practice new skills and apply new knowledge in combination with support, encouragement and feedback from colleagues, peers and superiors have been managed as particularly important in these studies. It seems that partner support can be an influence on the intention to stay, and the lack of partner support may represent a critical inhibitor of this intention and influences the IL.

Available Opportunities Outside the Organisation

Organisations can expect some wear and tear colleagues (Eddleston, 2009). It is also possible for workers to find differences between the career opportunities they may have in their companies and those available in the labour market; so that they may be inclined to pursue those lucrative and challenging opportunities, wherever they find them (Stahl et al., 2009). Moreover, a set of events or conditions associated with professional stagnation can contribute with people seeking better deals in other professional organisations.

Work–family Conflicts

A specific aspect of occupational or job stress is work–family conflict. It can be described as a form of conflict between work and family roles, in which the pressure of the key roles of work and family is mutually incompatible (Haar, Roche & Taylor, 2012).

Indeed, the work–family conflict as a topic has become very important in the research of organisational behaviour (Blomme et al., 2010a, 2010b). This relevance is caused by the social development that includes phenomena, such as, the inclusion of more women into the labour market and increasing couples that contribute to the family income, and as a result of this there are a greater number of hours devoted to paid work. Therefore, conflict can occur in two directions: the family interferes with work or work can interfere with family, in both cases, it results in a poor balance between family and work.

This work–family conflict is bidirectional, meaning that there is a mutual involvement. The work seems to have a disruptive effect on family life, and the family could have a negative effect on work demands (Mcnall et al., 2010). In the case of the hotel sector, the work–family conflicts appear to be an important reason for leaving work. The arguments are long working hours, low job security, irregular hours and low flexibility to coordinate exchange workdays.

Organisational Extrinsic Motivators

In general, motivators are all aspects of work that people consider important. Two main types of motivators are recognised: intrinsic and extrinsic motivators (Münderlein et al., 2013). The extrinsic and intrinsic motivators are a combination that is not always recognised separately, since a worker can be motivated by these two factors simultaneously.

Intrinsic motivators are given by the work itself or the learning opportunities it offers, that is, the worker is motivated by the satisfaction that is generated through the performance of work or learning it generates (Lai & Kapstad, 2009). The extrinsic aspects are composed of incentives that accompany the work, such as, receiving a salary, bonuses, prizes and awards (Münderlein et al., 2013).

Consequently, as motivators can increase the value of work, they can also increase attachment towards it. Lack of motivators can influence the IL or quit a job to find one containing the necessary motivation. Workers who enjoy their work and have motivators provided by the company will be less inclined to retire or leave the company voluntarily. On the contrary, the workers begin to worry about changing employers (Münderlein et al., 2013).

All of the above suggests a set of opportunities for advance studies of more conceptually complexity that allows one to understand and to explain the links between the causes of employee turnover through intention for leaving. Due to the above, in methodologically aspect, it recognises the need to implement and validate an instrument to measure the variables, space and job content, support perceived supervisor, support perceived peers, family–work conflict (FWC), available opportunities and organisational extrinsic motivators.

Therefore, to design an instrument that can measure the cause of IL and thereof, considering the characteristics of a specific sector, the following working hypotheses were proposed:

H₁: The items representative of dimensions will be interrelated, with Pearson coefficients greater than 0.500.

r_It–It ≥ 0.500

H₂: Variables within each dimension have an alpha equal to, or greater than, 0.500.

α_d ≥ 0.500

H₃: Variables within this instrument have an alpha equal to, or greater than, 0.700.

α_IMIS ≥ 0.700

Method

The design and construction of the pilot instrument took into account the literature review, as well as the references of the elements that influence the perception of the work environment and the IL; finally, the dimensions and items were set. It was framed as the Likert scales are with enough items relevant to assess the opinion, with statements containing no ambiguity and express approval or rejection of a given situation. Subjects respond to this questionnaire by picking a spot on the scale indicating the degree to define its position on the question, and the total index of attitudes was obtained by the summation of partial attitudes expressed in each answer.

The construction of this questionnaire with Likert scale (Briones, 1996) included the following steps:

Nominal definition of the attitude or variable.

Collection of items or indicators of that variable.

Determination of scores given to the categories of items.

Application of the interim scale to an appropriate sample, and calculation of individual scalar ratings.

Analysis of the items used to eliminate those inadequate.

Hierarchical categorisation of the scale.

Calculation of the validity and reliability of the scale.

Results

Nominal Definition of the Attitude or Variable

The design and development of the pilot instrument considered the literature review, the references of the elements that influence the perception of the work environment and the IL, and finally dimensions and items were set.

For the purpose of the study ‘causes of intention to leave’ is defined as the set of elements and external organisational characteristics that are perceived by its members and that affect their behaviour. This perception of workers in the sector can be direct or indirect. These perceptual characteristics are relatively permanent over time and are different from one organisation to another, and even between sections of the same company.

For measuring the causative or influencing factors of IL, an ordinal scale measure is proposed, using an instrument of 40 items, which corresponds to a frequency scale with five categories.

To carry out the fieldwork, it was contemplated to respect of integrity of individuals, and organisations maintaining anonymity and control of the results obtained, avoiding risks to research ethics.

The IL is by its nature an element caused by a set of multidimensional factors, so that their study can be very comprehensive. Therefore, it was narrowed in four dimensions that contain within themselves, other organisational and socio-psychological factors, which are also under review.

Because the relationship that it attempts to verify, between perceptions of the totality of individuals in organisations and the IL, one level for assessment was handled, and this was organisational.

The independent variables that were evaluated were: spaces and job content (SJC), perceived supervisory support, perceived support peers, FWC, available opportunities outside the organisation (AO) and extrinsic motivators: ‘intention to leave’ as the dependent variable.

Spaces and job content are defined as the degree to which the organisation provides adequate workspaces, as well as the degree to which employees see clear opportunities in their company for developing activities and tasks, in the way of their current organisation.

The PSS is understood as the degree to which employees receive effective support and guidance from their leaders during the course of their labour activities.

The perceived peer support (PPS) is understood as the degree to which the worker perceives the prevailing relations between members, characterised by: mutual respect, effective cooperation, solidarity and social integration.

The organisational extrinsic motivation (OEM) is defined as the degree to which the employee perceives that the organisation provides benefits that correspond to their contribution to the company or their performance.

The AO are assumed as the extent to which the worker perceives the existence of opportunities for growth and/or development in other institutions or organisations.

The FWC is understood as the degree to which the employee perceives the existence of conflicts between family and work, about effort and time dedicated to one factor or another.

The IL dependent variable is defined as the degree to which a worker perceives the need to leave the organisation.

Construction of Items

Fifty-three items were selected, using an ordinal scale of measurement. Each variable is allocated between three to seven positive and negative items, and it was intended that the selected items have the following characteristics:

Aim at the present.

Avoid giving too much information about the facts, or those that could be interpreted as such.

Not being ambiguous.

Be relevant to the attitude to be measured.

Represent favourable or unfavourable opinions.

Avoid using items with which everyone agrees or disagree with.

The items were formulated in a clear and simple language, taking care not to exceed 20 words, with a single logical sentence.

The positive and negative items are combined in a ratio of about 50 per cent to 50 per cent.

The 40 items are not exceeded.

Determination of Scores Given to the Categories of Items

The answers that the subjects can deliver to each statement are as follows: ‘strongly agree’, ‘agree’, ‘I’m not sure’, ‘disagree’, ‘strongly disagree’.

The scores for the items (as displayed in Table 1) involve a high or low frequency, depending on the individual agreement with the statement in the item.

To give meaning to the measurement, the table reflects different causes of IL, where a score is considered high or low depending on the number of items that sizes a variable. A high score indicates a high frequency, related to a possible organisational situation and a low score given indicates a low frequency, related to the lack of existence of a situation within the organisation.

Table 1.

Scores to Items

Scores to Items
	Totally Disagree	Disagree	Not Sure	Agree	Totally Agree
Positive items	1	2	3	4	5
Negative items	5	4	3	2	1

Source: Developed by the authors.

The variables studied as possible causes of IL are: spaces and content of job, with six items; perceived supervisory support, with six items; perceived support from peers, with six items; the work–family conflict, with three items; AO, with six items; and organisational extrinsic motivators, with six items. The IL as a dependent variable is dimensioned with seven items, so that the instrument has a total of 40 items.

The scores obtained of the total sum of items 4, 6, 14, 21, 22 and 33 of the instrument describe the extent to which people perceive SJC of the organisation as favourable for their performance.

SCJ low: 6–13 points

SCJ medium: 14–21 points

SCJ high: scores between 22 and 30

The scores obtained of the total sum of items 7, 11, 15, 20, 35 and 39 of the instrument describe the extent to which people PPS as favourable for their performance.

PPS low: 6–13 points

PPS medium: 14–21 points

PPS high: scores between 22 and 30

The score obtained of the total sum of items 3, 10, 23, 29, 34 and 40 of the instrument describe the extent to which people perceive PSS as favourable for their performance.

PSS low: 6–13 points

PSS medium: 14–21 points

PSS high: scores between 22 and 30

The scores obtained of the total sum of the items 5, 16, 24, 28, 30 and 38 of the instrument describe the extent to which people perceive OEM and favourable performance.

OEM low: 6–13 points

OEM medium: 14–21 points

OEM high: scores between 22 and 30

The scores obtained of the total sum of items 2, 12, 17, 25, 27 and 31 of the instrument describe the extent to which people perceive the AO as motivation for additional activities or alternative to work.

AO low: 6–13 points

AO medium: 14–21 points

AO high: scores between 22 and 30

The scores obtained of the total sum of the items 8, 19 and 36 of the instrument describe the extent to which people perceive family–work or work–family gaps (FWC) as conflict to realise work activities.

FWC low: 3–6 points

FWC medium: 7–10 points

FWC high: scores between 11 and 15

The scores obtained of the total sum of items 1, 9, 13, 18, 26, 32 and 37 of the instrument describe the degree to which individuals have the IL in the organisation.

IL low: 7–15 points

IL medium: 16–25 points

IL high: scores between 26 and 35

See Table 2 for better appreciation of the interpretation of the scores.

Validation

The questionnaire was subjected to validation of content in a focus group consisting of 12 expert judges (organisational psychologists from different universities and human resource (HR) managers from the hospitality industry), who assessed the instrument using a format that included the variables to be assessed, their items, and the following evaluation criteria:

Relevance of the item

Consistency of the item

Clarity of item

Sufficiency of items to assess each dimension

Table 2.

Scores Table

Dimension	Score
	Low	Medium	High
SCJ	6–13	14–21	22–30
PPS	6–13	14–21	22–30
PSS	6–13	14–21	22–30
OEM	6–13	14–21	22–30
OA	6–13	14–21	22–30
CFW	3–6	7–10	11–15
IL	7–15	16–25	26–35
	Score range

Source: Developed by the authors.

The judges were asked to conduct a validation of the instrument by presenting a preliminary list of statements that would be relevant to measure the degree of favourability/disfavourability of individuals to certain object. The judges, in this case, did not act as subjects who responded with their opinions about the statements or items, but they conducted a judgement of favourability. That is, they did not express their opinion on the statement that was presented to them, but they expressed the degree of favourability that statements expressed regarding dimensions, in terms of the evaluation criteria (pertinence, relevance, consistency and clarity), marking, as shown in Figure 1, the value in the corresponding box being interpreted as totally unfavourable as being the number 1 and the entirely favourable being represented as number 5 in the polarised scale.

Figure 1.

Source: Authors’ own.

The judges were asked to conduct a validation of the instrument and were presented a preliminary list of statements that were thought might be relevant to measure the degree of favourability/disfavourability of individuals towards certain objects. In this case, the judges did not act as subjects who responded with their opinions about the statements or items, but they conducted a judgement of favourability. That is, they did not express their opinion on the statement that was presented to them, but the degree of favourability that statements expressed regarding dimensions, in terms of the evaluation criteria (pertinence, relevance, consistency and clarity), marking, as shown in Figure 1, the value in the corresponding box being interpreted as totally unfavourable being the number 1, and entirely favourable represented as number 1 in a polarised scale.

This assessment took place over a period of 20 days, after which, formats were analysed, and based on them corrections were carried out, taking into account the suggestions of the judges. Initially, the instrument consisted of 53 items, of which 13 were eliminated due to the lack of relevance of items, the way of writing, by the similarity of content or the lack of meaning and clarity.

Calculation of Reliability

The sample to test the instrument consisted of 145 persons occupationally linked to hotel companies in Puerto Vallarta (Mexico). The instrument was applied three times to make adjustments to achieve the reliability shown. The respondent took 15 minutes to fill the questionnaire.

The reliability of the instrument was evaluated through split-halves technique with a single application of the instrument. This instrument had 40 items and other 40 that were repeated, forming a total of 80 items. Items were evaluated for his mirror with coefficient correlation ‘r’ Pearson. Cronbach’s alpha coefficient was calculated in the different components of the instrument reflecting the degree to which it covaries items, constitutes the test, as an indicator of internal consistency. The procedure was performed using SPSS.

Reliability

The first phase of designing the instrument for measuring the IL is carried out based on the technique of split halves—which are considered the 40 items with a mirror thereof forming an instrument of 80 items. In this instance of scanning, the index Pearson correlation is used, between items as evidence to face validity. The intention was to find that respondents interpret the same concepts included in the items, correlating mirrors and looking valid item in the process towards the concept validity and construct validity. In the future, one may use confirmatory factorial analysis (CFA) to get more evidence of validity construct.

For reliability, determination by Cronbach’s alpha has been used, which allows observing the extent to which items covariate that constitutes the test, as evidence of internal consistency.

As shown in Table 3, the correlation coefficients obtained in the 40 items are acceptable, with some exceptions, such as, item 15 (PPS), 16 (OEM), 22 (SJC) and 27 (AO) all concerning to the causal of IL, but 38 of the 40 are significant at 0.01 and only two at the 0.05 level.

Table 3.

Correlations between Items

Pearson Correlations Item
	41		51		61		71
1	0.572**	11	0.700**	21	0.657**	31	0.748**
	42		52		62		72
2	0.559**	12	0.775**	22	0.404**	32	0.862**
	43		53		63		73
3	0.536**	13	0.530**	23	0.990**	33	0.823**
	44		54		64		74
4	0.606**	14	0.522**	24	0.866**	34	0.949**
	45		55		65		75
5	0.646**	15	0.263*	25	0.610**	35	0.711**
	46		56		66		76
6	0.866**	16	0.407**	26	0.891**	36	0.647**
	47		57		67		77
7	0.739**	17	0.515**	27	0.362*	37	0.769**
	48		58		68		78
8	0.529**	18	0.981**	28	0.846**	38	0.732**
	49		59		69		79
9	0.746**	19	0.590**	29	0.944**	39	0.763**
	50		60		70		80
10	0.848**	20	0.976**	30	0.641**	40	0.769**

Source: Developed by the authors.

Note: *Correlation is significant at level 0.05 (two-tailed); **correlation is significant at level 0.01 (two-tailed).

Table 4.

Alpha Coefficients for Dimension

Dimensions	Alpha
SCJ	0.702
PPS	0.652
PSS	0.736
OEM	0.866
AO	0.533
CFW	0.470
IL	0.831
IMIS	0.892

Source: Developed by the authors.

As seen in Table 4, in the SJC, alpha is 0.702, in the PSS it is 0.736, in the PPS is 0.652, in the OEM it is 0.866, in the AO it is 0.533 and in the FWC it is 0.470, and an alpha of 0.831 for the ‘IL’ with an alpha of 0.892 contained for all the instruments (IMIS, Instrument of Measuring of Intention to Leave).

Discussion

The design of an instrument for measuring the IL (IMIS) in a specific sector, such as, the hotels in Puerto Vallarta, Mexico, allows to determine interactions between some variables and shows the need to incorporate contextual and cultural variables for the analysis. The power of the instrument lies in the possibility of revealing interactions, but above all, to reveal the absence of relationships or direct interactions, reinforcing the evidence about the complexity of the phenomenon, and the major limitations of studies that looks to generalise the causal relationships between variables or types of organisations.

First, it should be noted that the instrument met the goal of conceptually relating the main variables, the causal of IL, that are relating to the organisation and its environment within a theoretical framework and an empirical framework as an exploratory support for the consolidation of a measuring instrument in larger and more heterogeneous samples.

The instrument proved to be reliable as a whole and in all dimensions in preliminary analysis. Thus, the results of correlated items revealed through the instrument measure the dimensions of causes explaining IL, and their respective variables. And the results of the alpha coefficient showed that the instrument is consistent in its dimensions and as a whole.

In conclusion, it is argued that IMIS is a validity and reliable instrument to characterise the IL in an organisation. In this first instance, certain solidity is demonstrated, although continuous refining, seeking to be useful to operationalise the IL of any organisation.

However, the results indicate that the IL, in fact, can be characterised using a set of variables that interact with other variables of the subjects, the environment and the organisation itself. In this study, it was expected to find that the total set of relationships in the item–item test were above 0.500 in the correlation coefficient Pearson, since it was assumed that attitudes were the same to the same statement. However, it appears that 36 items of 40 were above 0.500 only four were below 0.500. This could imply that if the wording of these four items fits, it could create an instrument with greater reliability.

Moreover, it was expected that all of the dimensions in the instrument were above 0.600 in the Cronbach’s alpha. However, and despite the total alpha being 0.892, two dimensions were observed below 0.600, as in the case of opportunities available outside the organisation and work–family conflict, this dimension being the lowest at 0.470. This can be attributed to the amount of items used (only three), which may suggest increasing the number of items to be used in the instrument. Generally, five dimensions of seven had an alpha above 0.600.

While the sample is significant for the sector, the results show that for future research it would be necessary to include variables with clear content that could account for aspects and specific features that are concrete of organisational cultures and individuals. This would give greater explanatory power to the instrument and allow its adaptation so as to study the IL from the same industry, but in other contexts, and also to investigate the phenomenon in other industries in the same context.

In addition, to contemplate attributes which are more specifically cultural and contextual type, it is suggested for the need for research with designs that enable the integration of qualitative and quantitative approaches and techniques to build hypotheses and assumptions to explore and interpret other emerging dimensions of the phenomenon in the organisational groups and individuals.

The limitations of the instrument are, of course, that the sample is small with only 145 cases and only of hotel companies being included. Moreover, the lack of a factors analysis by ‘R’ type components, to check solidly the validity of the construct of full scale. However, it is believed that the results contribute a basis for other studies to confirm the instrument in a second instance in hotel samples in national and Latin American organisations.

Footnotes

Acknowledgements

This research was supported in part by Promep, Professors Improvement Programme. We are grateful for the support received from our colleague Germán Elías Jiménez Montalbán for their comments and translation. The authors would like to gratefully acknowledge the editor, the associate editor and the anonymous reviewers for their comments and suggestions on the article.

References

Adams

G.A.

, & Beehr

T.A.

(1998). Turnover and retirement: A comparison of their similarities and differences. Personnel Psychology, 51(3), 643–665.

Allen

D.G.

, & Griffeth

R.W.

(1999). Job performance and turnover: A review and integrative multi-route model. Human Resource Management Review, 9(3), 525–548.

Biron

, & Boon

(2013). Performance and turnover intentions: A social exchange perspective. Journal of Managerial Psychology, 28(5), 511–531.

Blomme

R.J.

, Rheede

A.V.

, & Tromp

D.M.

(2010a). Work-family conflict as a cause for turnover intentions in the hospitality industry. Tourism and Hospitality Research, 10(4), 269–285.

Blomme

R.J.

, Rheede

A.V.

, & Tromp

D.M.

(2010b). The use of the psychological contract to explain turnover intentions in the hospitality industry: A research study on the impact of gender on the turnover intentions of highly educated employees. The International Journal of Human Resource Management, 21(1), 144–162.

Briones

(1996). Methodology of quantitative research in the social sciences. Bogotá, Colombia: ICFES.

Bycio

, Hackett

R.D.

, & Alvares

K.M.

(1990). Job performance and turnover: A review and meta-analysis. Applied Psychology: An International Review, 39(1), 47–76.

Cheng

, & Waldenberger

(2013). Does training affect individuals’ turnover intention? Evidence from China. Journal of Chinese Human Resource, 4(1), 16–38.

Chinomona

, & Chinomona

(2013). The influence of employees’ perceptions of organisational politics on turnover intentions in Zimbabwe’s SME sector. South African Journal of Business Management, 44(2), 57–67.

10.

Deloitte. (2010). Hospitality 2015: Game changers or spectators. London: Deloitte LLP.

11.

Denvir

, & McMahon

(1992). Labour turnover in London hotels and the cost effectiveness of preventative measures. International Journal of Hospitality Management, 11(2), 143–154.

12.

Domínguez

, Roncancio

, Marcelino-Aranda, & Ramírez

(2015). Instrument measuring intentions to leave. Asia Pacific Journal of Management Research and Innovation, 11(4), 1–10.

13.

Dusek

G.A.

, Ruppel

C.P.

, Yurova

, & Clarke

(2014). The role of employee services orientation in turnover in the U.S. hotel industry. Journal of Organisational Culture, Comunications and Conflict, 18(2), 87–105.

14.

Eddleston

K.A.

(2009). The effects of social comparisons on managerial career satisfaction and turnover intentions. Career Development International, 14(1), 87–110.

15.

Gamor

, Amissah

E.F.

, & Boakye

K.A.

(2014). Work-family conflict among hotel employees in Sekondi-Takoradi Metropolis, Ghana. Tourism Management Perspectives, 12, 1–8.

16.

Griffeth

R.W.

, Hom

P.W.

, & Gaertner

(2000). A meta-analysis of antecedents and correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26(3), 463–488.

17.

Haar

J.M.

, Roche

, & Taylor

(2012). Work–family conflict and turnover intentions of indigenous employees: The importance of the whanau/family for Maori. The International Journal of Human Resource Management, 23(12), 2546–2560.

18.

Jackofsky

(1984). Turnover and job performance: An integrated process model. Academy of Management Review, 9(1), 74–83.

19.

Karatepe

O.M.

(2012). The effects of coworker and perceived organisational support on hotel employee outcomes: The moderating role of job embeddedness. Journal of Hospitality and Tourism Research, 36(4), 495–516.

20.

Kerlinger

F.N.

, & Lee

H.B.

(2002). Investigación del comportamiento: Métodos de investigación en ciencias sociales. México: Mc Graw Hill editores.

21.

Kyei-Poku

I.A.

, & Miller

(2013). Impact of employee merger satisfaction on organisational commitment and turnover intentions: A study of a Canadian financial institution. International Journal of Management, 30(4), 205–224.

22.

Lai

, & Kapstad

J.C.

(2009). Perceived competence mobilisation: An explorative study of predictors and impact on turnover intentions. The International Journal of Human Resource Management, 20(9), 1985–1998.

23.

, & Jones

C.B.

(2012). A literature review of nursing turnover costs. Journal of Nursing Management, 21(3), 405–418.

24.

Liu

, Cai

, Li

, Shi

, & Fang

(2013). Leadership style and employee turnover intentions: A social identity perspective. Career Development International, 18(3), 305–324.

25.

Mcnall

L.A.

, Masuda

A.D.

, & Nicklin

J.M.

(2010). Flexible work arrangements, job satisfaction, and turnover intentions: The mediating role of work-to-family enrichment. The Journal of Psychology, 144(1), 61–81.

26.

Mohsin

, Lengler

, & Aguzzoli

(2015). Staff turnover in hotels: Exploring the quadratic and linear relationships. Tourism Management, 51, 35–48.

27.

Münderlein

, Ybema

J.F.

, & Koster

(2013). Happily ever after? Explaining turnover and retirement intentions of older workers in the Netherlands. Career Development International, 18(6), 548–568.

28.

Pizam

, & Thornburg

S.W.

(2000). Absenteeism and voluntary turnover in Central Florida hotels: A pilot study. International Journal of Hospitality Management, 19(2), 211–217.

29.

Regts

, & Molleman

(2012). To leave or not to leave: When receiving interpersonal citizenship behavior influences an employee’s turnover intention. Human Relations The Tavistock Institute, 66(193), 192–218.

30.

Schmidt

J.A.

, & Lee

(2008). Voluntary retirement and organisational turnover intentions: The differential associations with work and non-work commitment constructs. Journal of Business Psychology, 22(4), 297–309.

31.

Siebert

W.S.

, & Zubanov

(2009). Searching for the optimal level of employee turnover: A study of a large U.K. retail organisation. Academy of Management Journal, 52(2), 294–313.

32.

Stahl

G.K.

, Chua

C.H.

, Caligiuri

, Cerdin

J.-L.

, & Taniguchi

(2009). Predictors of turnover intentions in learning-driven and demand-driven international assigments: The role of repatriation concerns, satisfaction with company support, and perceived career advancement opportunities. Human Resource Management, 48(1), 89–109.

33.

Staufenbiel

, & König

C.J.

(2010). A model for the effects of job insecurity on performance, turnover intention, and absenteeism. Journal of Occupational and Organisational Psychology, 83(1), 101–117.

34.

Steers

R.M.

, & Mowday

R.T.

(1981). Employee turnover and post-decision accommodation processes. Research in Organisational Behavior, 3, 235–281.

35.

Van Breukelen

, Van der Vlist

, & Steensma

(2004). Oluntary employee turnover: Combining variables from the ‘traditional’ turnover literature with the theory of planned behavior. Journal of Organisational Behavior, 25(7), 893–914.

36.

Wright

T.A.

, & Bonett

D.G.

(2007). Job satisfaction and psychological well-being as nonadditive predictors of workplace turnover. Journal of Management, 33(2), 141–160.

37.

Yang

, Gong

, & Huo

(2011). Proactive personality, social capital, helping, and turnover intentions. Journal of Managerial Psychology, 26(8), 739–760.

38.

Zimmerman

R.D.

, & Darnold

T.C.

(2009). The impact of job performance on employee turnover intentions and the voluntary turnover process: A meta-analysis and path model. Personnel Review, 38(2), 142–158.