This chapter will provide insight about the research framework, the hypotheses, the research procedure, data collection the sample, the measurements as well as the data analysis methods.
Research Framework
Figure 3.1 illustrates the research framework which demonstrates the effect of perceived supervisor support, impression management motives and choice of contract on OCB. It also shows the moderating effect of time remaining on the contract as well as expectation of contract continuation on the relationship between some of the independent variables and OCB.
There are three independent variables which are perceived supervisor support, impression management motives, and choice of contract. The moderators are time remaining on the contract and expectation of contract continuation. The dependent variable is organizational citizenship behavior. All of the variables used in this study have been already proposed in literature and the researcher tried as much as possible to use instruments that have already been tested as reliable and valid.
Figure 3.1. Research Framework
Time remaining on contract Perceived
supervisor support
Organizational Citizenship
Behavior Expectation of
contract continuation
Impression management
motives
Choice of contract
H5 + H1 +
H2 +
H3 -
H4 -
28
Research Hypotheses
Hypothesis 1: Perceived supervisor support has a positive relationship with contingent workers’
OCB in the nonprofit sector in Haiti.
Hypothesis 1a: Perceived supervisor support has a positive relationship with contingent
workers’ altruism in the nonprofit sector in Haiti
Hypothesis 1b: Perceived supervisor support has a positive relationship with contingent
workers’ conscientiousness behaviors in the non-profit sector in Haiti.
Hypothesis 1c: Perceived supervisor support has a positive relationship with contingent
workers’ attendance behaviors in the nonprofit sector in Haiti.
Hypothesis 2: Impression management motives have a positive relationship with OCB of
contingent workers in the non-profit sector in Haiti.
Hypothesis 2a: Impression management motives have a positive relationship with the altruism of
contingent workers in the non-profit sector in Haiti.
Hypothesis 2b: Impression management motives have a positive relationship with
conscientiousness behaviors of contingent workers in the non-profit sector in Haiti.
Hypothesis 2c: Impression management motives have a positive relationship with attendance
behaviors of contingent workers in the non-profit sector in Haiti.
Hypothesis 3: Involuntary choice of contract has a negative relationship with OCB of
contingent workers in the non-profit sector in Haiti.
Hypothesis 3a: Involuntary choice of contract has a negative relationship with altruism of
contingent workers in the non-profit sector in Haiti.
Hypothesis 3b: Involuntary choice of contract has a negative relationship with
conscientiousness of contingent workers in the non-profit sector in Haiti.
Hypothesis 3c: Involuntary choice of contract has a negative relationship with attendance of contingent workers in the non-profit sector in Haiti.
29
Hypothesis 4: Time remaining on contract negatively moderates the relationship between IMM and OCB with shorter time remaining strengthening the relationship between IM motives and the OCB of contingent workers in the non-profit sector in Haiti.
Hypothesis 4a: Time remaining on contract negatively moderates the relationship between IMM and altruism with shorter time remaining strengthening the relationship between IM motives and the altruism of contingent workers in the non-profit sector in Haiti.
Hypothesis 4b: Time remaining on contract negatively moderates the relationship between IMM and conscientiousness with shorter time remaining strengthening the relationship between IM motives and the conscientiousness of contingent workers in the non-profit sector in Haiti.
Hypothesis 4c: Time remaining on contract negatively moderates the relationship between IMM and attendance with shorter time remaining strengthening the relationship between IM motives and the attendance of contingent workers in the non-profit sector in Haiti.
Hypothesis 5: Expectation of contract continuation negatively moderates the relationship between IMM and OCB with negative expectation strengthening the relationship between Impression management motives and OCB
Hypothesis 5a: Expectation of contract continuation negatively moderates the relationship between IMM and altruism with negative expectation strengthening the relationship between impression management motives and altruism.
Hypothesis 5b: Expectation of contract continuation negatively moderates the relationship between IMM and conscientiousness with negative expectation strengthening the relationship between impression management motives and conscientiousness.
Hypothesis 5c: Expectation of contract continuation negatively moderates the relationship between IMM and attendance with negative expectation strengthening the relationship between impression management motives and attendance.
30
Research Procedure
This section explains the steps taken in order to develop this research. The procedure is as shown in figure 3.2. The researcher considered an exploratory study already done in the subject as a lead for the research questions. In order to come up with viable research questions, the second step for this study was a review of literature to develop the basis for the research. The findings from previous studies allowed the identification and a better framing for the research questions. The research questions and hypotheses were developed from literature review. Once the research variables have been clearly identified, a framework was developed in order to represent the relationships between them. From literature the researcher chose the measures to collect the data needed in order to answer the research questions and test the hypotheses. A pilot study was conducted to test initial reliability and validity of the questionnaire. After gathering data, a series of analysis was performed in order to answer the research questions. Afterwards, the researcher discussed the results, answered the research questions, suggested future research and concluded the study.
Figure 3.2. Research Procedure
Identification of the research topic Literature Review
Formulation of research questions and hypotheses
Analysis of the data Research framework
Pilot study
Selection and design of research instrument
Collection of data
Conclusions and suggestions
31
Research Design Sampling
The researcher opted for a quantitative method through distribution of survey questionnaires to measure the impact of perceived supervisor support, impression management motives, and choice of contract on OCB of contingent workers in Haiti. Survey questionnaires were distributed by the researcher to the targeted contingent workers in Haiti. Initial contact was made with human resource representatives of organizations in the non-profit sector explaining the purpose of the study via email or telephone in order for the researcher to be granted access to the potential respondents. In order to ensure validity and reliability, a pre-test on a similar sample with the French version of the questionnaire was performed with an electronic version of the final questionnaire. The researcher therefore went to various non-profit organizations in order to find participants and collected some data immediately after the participant finished filling out the questionnaire. The participants were contingent workers in the non-profit sector in Port-au-Prince which is the capital of Haiti and seven regions of the country.
Because there are different types of contingent workers and they do not all share the same attitudes and behavioral outcomes, this study focuses on workers not hired through an agency but by the organization itself. They are also called direct-hires or in-house temporaries. This sampling was done according to Wilkin's (2013) and Marler et al. (2002) studies which suggest that future studies should specify exactly the type of contingent workers being studied in order to have a more uniform sample and clearer understanding of a specific group of workers and better results for cross-cultural comparisons. Therefore, the criteria for inclusion in this study were the following:
1. Participants must be contingent workers and be under the direct supervision of the organization which means no intermediary agency must be involved in the employment relationship.
2. The participants must be working under contract with precise start and end dates.
3. The participants must be employed in the non-profit sector
32
A pilot study was conducted on a similar sample of 44 contingent workers in the non-profit sector in Haiti. The different variables yield satisfactory Cronbach`s Alpha for the variables being studied.
Data Collection
The original questionnaire was translated into French which is an official language of the country then a backward translation into English was performed in order for the researcher to be sure that there is no major loss of meaning for the items. Suggestions were made by an HR specialist to leave the English version of the items on OCB as there were some supervisors who were not locals therefore not necessarily French literate. Consequently, the questionnaire for supervisors contained both French and English items for OCB. The pilot test was conducted on 44 contingent workers in different non-profit organizations.
To ensure confidentiality of the research, questionnaires were provided in an envelope containing both the supervisor and the subordinate questionnaire. The researcher went directly to the organizations who granted access in order to explain the coding process so that confidentiality could be maintained and to reduce biases. The supervisor, with a list, coded each questionnaire and gave the subordinate the part he/she had to fill. Both supervisor and subordinate had the same code on their survey. The researcher collected some surveys the same day and also returned a few weeks later in order to collect the remaining ones. They were individually put in a sealed envelope then regrouped in a bigger case by department in order to facilitate collection by the researcher. The researcher ensured each time a bag was collected to write on the envelope a code randomly created in order not to mix the different organizations.
Sample Profile
In this research, 270 questionnaires were distributed in 9 different organizations, and, with a return rate of 67%, 181 questionnaires were used for this study.
From the 181 respondents, 82 are from one organization executing a communication project.
The majority of the sample (88, or 48%) had a bachelor degree. There were 86 (47%) females, 95 (52%) males and ages ranging from 23-32 years had the highest frequency (92, or 50%) and most of the respondents (70, or 38%) have been working in their organization for more than a
33
year up to three years. In this sample, 89 (49%) had mostly no temporary work experience prior to joining their actual organization and the current contract duration of the majority (82, or 45%) was of less than two months.
The respondents had a minimum of one day and a maximum of 360 days before contract expires with a mean of 74.56 days. Their contracts have been renewed from a minimum of zero time to 23 times with a mean of 8.8 times. The majority (82 or 45%) is currently employed in an organization providing communication services. Table 3.1 shows the descriptive statistics of the sample for this study.
Table 3.1.
Descriptive Statistics of the Sample (N=181)
Variable Item Frequency Percentage (%)
Age Less than 23 4 2.2
23-32 92 50.8
33-42 74 40.9
43-52 8 4.4
53-62 3 1.7
Gender Female 86 47.5
Male 95 52.5
Education High school diploma 59 32.6
Bachelor degree 88 48.6
Master degree 10 5.5
Doctorate 1 .6
Other 23 12.7
(continued)
34 Table 3.1. (continued)
Variable Item Frequency Percentage (%)
Tenure Less than three months 8 4.4
Three to six months 7 3.9
Seven to 12 months 10 5.5
More than a year up to three years 70 38.7 More than three up to five years 37 20.4 More than five up to seven years 39 21.5
More than seven years 10 5.5
Contract duration Less than two months 82 45.3
Two to four months 32 17.7
Five to seven months 25 13.8
Eight to 10 months 5 2.8
11 to 13 months 25 13.8
Over 13 months 12 6.6
Previous temporary work experience
Zero 89 49.2
Less than three months 15 8.3
Three to six months 14 7.7
Seven to 12 months 21 11.6
More than a year up to three years 18 9.9 More than three years up to five
years 18 9.9
More than five up to seven years 4 2.2
More than seven years 2 1.1
(continued)
35 Table 3.1. (continued)
Variable Item Frequency Percentage (%)
Organization nationality
Other 12 6.6
Local 16 8.8
International 165 91.2
Expectation of contract continuation
Yes 153 84.5
No 28 15.5
Organization Type Health 11 6.1
Education 2 1.1
Religious 5 2.8
Social Service 12 6.6
Humanitarian 31 17.1
Environment 7 3.9
Housing 2 1.1
Communication 82 45.3
Humanitarian_Environment_Socia
l service_Housing 17 9.4
Measurement
A survey questionnaire was developed for the collection of data for this study. The variables, as identified from literature were measured with items already used for that purpose. The questionnaire was divided into five parts and 7-point Likert scales were provided as options used by the participants to score their answers. Please refer to Appendix E and F for the complete survey questionnaire.
36
Impression Management Motives
Impression management motives (IMM) relate to people who are motivated to manage impressions in order to avoid projecting a negative image and to obtain rewards (Rioux &
Penner, 2001). In that section of the questionnaire, a definition of OCB as well as some examples was provided. Respondents were asked, how important, according to the statements relating to impression management motives, it was for them to engage in OCB. The impression management motives were measured using the citizenship motives’ impression management items developed by Rioux and Penner (2001). It contained six items and sample items are: “to impress my coworkers” as acquisitive motive and “to avoid looking lazy” as a defensive motive with response options ranging from 1 (not at all important) to 7 (very important). The item
“Because I want a raise” was replaced by “Because I want to be noticed for future work opportunities” in order to better reflect the situation of the sample. The higher the score is, the higher the impression management motives.
Perceived Supervisor Support
Perceived supervisor support refers to the belief the employee has that the supervisor acknowledges hard work from the employee and shows genuine attention to ensure the employee’s wellbeing (Kottke & Sharafinski, 1988). The measure for perceived supervisor support was measured with 12 items using the PSS scale developed by Kottke and Sharafinski (1988) which consisted mostly in replacing the word organization by the word supervisor. The psychometric properties of the scale were demonstrated by the developers and the items showed the same factor structure as the perceived organizational support scale while there was low correlation between the two scales showing very little overlapping. The response options for the items ranged from 1 (strongly disagree) to 7 (strongly agree). A sample item is “My supervisor wants to know if I have any complaints”.
37
Choice of Contract
The choice of contract refers to the involuntary reasons for working under temporary contract.
The choice of contract was measured with six items measuring involuntary reasons for working under contract from the scale developed by Tan and Tan (2002) with reference to the eight factors highlighted by the study of Ellingson, Gruys, and Sackett (1998). Participants rated their answers from 1 (strongly disagree) to 7 (strongly agree) with high levels representing involuntary reasons and low levels expressing a more voluntary choice. A sample item is “I am working under contract because it is difficult to find a permanent job that is satisfactory”.
Organizational Citizenship Behavior
Organizational Citizenship Behavior (OCB) refers to these behaviors that are individual, discretionary, and not directly or formally recognized by the reward system, but in the aggregate contribute to the effectiveness of the organization (Organ, 1988). Twelve items from the scale developed by Smith et al. (1983) were used to measure organizational citizenship behavior illustrating the behaviors oriented towards individuals (altruism) and towards the organization (conscientiousness and attendance). Sample items for altruism and conscientiousness are “help others who have heavy workloads” and “does not take extra breaks”. A sample of attendance behavior is “gives advance notice if unable to come to work”. Supervisors were asked to rate how frequently their subordinate engaged in OCB with a scale ranging from 1 (never) to 7 (always).
Temporal Factors
The temporal factors considered for this study refer to variables relating to time remaining on contract and future expectations of employment which may influence the contingent worker’s outcomes.
38 Expectation of contract continuation.
Expectation of contract continuation is assessed as a dichotomous variable. The respondent has to answer yes or no to the following question: Do you think your present employment contract will be renewed when it expires?
Time remaining on contract.
Time remaining on contract is assessed as a continuous variable where the respondent has to write the time remaining on contract in months, in weeks in case there is less than a month and in days in case there is less than a week left.
Demographics
Participants were asked to provide their age, education, gender, tenure in the current organization, type of organization, their contract duration, number of times contract was renewed, and whether they had previous temporary work experience before joining their current organization or not in order to better understand their profile and the impact these may have on the study variables.
Control Variables
Past research has demonstrated that age, educational level, gender and tenure do impact employee’s engagement in OCB (Podsakoff et al., 2000). Job satisfaction is considered as being the main predictor of OCB, therefore a general measure of job satisfaction from Wright and Cropanzano (1998) in the questionnaire helped control for this factor. It had five items with a sample being: “All in all, how satisfied are you with your salary?” Respondents had answer options from 1 (very dissatisfied) to 7 (very satisfied).
39
The demographic variables included in this study and which were also used as control variables are: gender assessed as a dichotomous variable; age, educational level, and tenure assessed as ordinal variables.
A summary of the items used in the questionnaire as well as sources is provided in table 3.2.
Table 3.2.
Measurement Scale Items
Variable Items References
Perceived supervisor support
12 items Kottke and Sharafinski (1988)
Impression
management motives
6 items Rioux and Penner (2001)
Choice of contract 6 items Ellingson, Gruys, and Sackett(1998), Tan and Tan (2002)
Organizational citizenship behaviors
12 items
Smith et al., (1983)
Altruism 7 items
Conscientiousness 3 items
Attendance 2 items
Validity and Reliability
Reliability and validity refer to two concepts attempting to ensure that instruments are stable overtime and effectively measure what they are intended to. Albeit the instruments used for this study have been previously validated, it was important for the researcher to some extent test content validity by having expert review of the questionnaire and by conducting a pretest on a similar sample in order to make necessary modifications and subsequently enhance the questionnaire and ensure better reliability
40
Exploratory Factor Analysis
The SPSS software was used in order to do the preliminary exploration of the data in order to check for errors, outliers, and the distribution of the data. The whole questionnaire was subjected to factor analysis to see how the different measurements of the current study matched the original factor structure of the measurement scales. It was important to conduct Kaiser-Mayer-Olkin (KMO) and Bartlett’s Test of Sphericity on the data in order to ensure suitability for exploratory factor analysis (Hair, Anderson, Tatham, & Black, 1998). The KMO result for the exploratory factor analysis is .834; the Barlett’s Test had a significance of .000. Factors with an eigenvalue larger than one were used. Seven factors were extracted with a total variance explained of 65.046 and varimax rotation method was used (Kinnear & Gray, 2000).
The items loaded mostly on their intended components with OCB loading on three stepping stone reasons. Table 3.3 shows the final EFA table for all the variables with the factor loading for each item. The first component relates to PSS, the second altruism, the third IMM, the fourth conscientiousness dimension of OCB, the fifth and the sixth, stepping stone and involuntary choice of contract and the last represents attendance dimension of OCB.
Table 3.3.
Rotated Component Matrix for all Variables
Component
1 2 3 4 5 6 7
PSS7_My supervisor really cares about my well-being
41 Table 3.3. (continued)
Component
1 2 3 4 5 6 7
PSS11_My supervisor shows a lot of concern for me
.801
PSS10_My supervisor cares about my general satisfaction at work
.778 PSS5_My supervisor takes my best interests
into account when he/she makes decisions that affect me
.773
PSS6_Help is available from my supervisor when I have a problem
.763
PSS13_My supervisor takes pride in my accomplishments
.753
PSS8_If I did the best job possible my supervisor would be sure to notice
.715
PSS4_My supervisor wants to know if I have any complaints
.675
PSS9_My supervisor is willing to help me when I need a special favor
.669 OCB3_Orients new people even though it is
not required
.843
OCB2_Volunteers to help for things that are not required
IMM9_Because I want to be considered for
IMM9_Because I want to be considered for