Union Politicization in Sri Lanka : Dimensions and Measurement

Through trade unions are mostly viewed as employee organizations protecting employee‟s interests, they have political dimension too. Politicization of employee unions has added an extra complexity not only to their behavior, but also to their impact on organizations in industrial relations. Therefore, this political dimension of employee unions has to be taken into account in studies on unions. Researchers have identified two dimensions of union politicization namely political instrumentality and political ideology. Measurement of union politicization should incorporate both the political instrumentality and political ideology dimensions of union politicization. However, past researches have used either the political instrumentality or political ideology in measuring union politicization depending on their research context. The aim of this study is to test the goodness of measure of the union politicization in Sri Lankan Context. Dhammika, K.A.S., Ahmad, F.B., Sam, T.L. Wayamba Journal of Management 3 (2) 61 Data was collected from 136 public sector employee selected using stratified random sampling. Exploratory factor analysis and reliability testing were applied for testing the validity and reliability of the measurement. It was found that both political instrumentality and political belief dimensions are validly and reliably held in explaining the union politicization in Sri Lankan context. Further, results and implications are discussed so that it can be used for future research on union politicization. Key Wards: Union Politicization, Political Instrumentality, Political Belief and Measurement


Introduction
Unions as employee organizations and employees" behavior as union members have been subjective to an extensive research over the past decades (Parks, Gallagher, and Fullagar, 1995).It is due to the fact that employee unions bring important implications not only to the industrial relation climate of organizations, but also to the overall functioning of them (Wickens, 2008).As a result, unions have been studied with different perspectives such as economic, psychological, political, and managerial perspectives (Gani, 1996;Opatha, 2001;Biyanwilla, 2003).But union research in recent years has placed an increased emphasis on organizational and behavioral approach to the study of employee unions (Parks, et al., 1995).Due to this renewal interest, some behavioral aspects related to unions, have been extensively explored.Among such variables, union commitment (Snape, Redman, & Chan, 2000), union loyalty and union satisfaction, (Iverson & Kuruvilla, 1995), union participation (Fuller & Hester, 1998) and pro union attitudes and union instrumentality (Chan, Tong-Qing, Redman and Snape, 2006) have been explored.However, it is noted that some contextual aspects related to employee unions have not been much studied.Among such contextual dimention of unions not adequaltely examined , union ploiticization stands prominent given the fact that unions are highly political prone in some contexts.Unions in South Asian regoins, and in Sri Lanka particularly are identified as highly politicized (Biyanwilla, 2003;Jinadasa & Opatha, 1999).Therefore, this political dimension of employee unions has to be taken in account in studies on union behaviour.However, it seems that the research on this important aspect is lagging behind specially in Sri lankan context.This may be attributed to non avilability of empiricaly tested measurement instrument for union politicizaion.Therefore, this study was purposed to test the dimentionality of union politicization in Sri lankan context

2.Union Politics
Though trade unions are conceptually viewed as employee organizations which represents the employees" interests, researcher claimed that they have a political dimension too.Hyman & Gumbrell-McCormick, (2010) proposed that trade unions are not merely economic actors, they are necessarily actors in the political arena.This political dimension of employee unions has added an extra complexity not only to their behavior but also to their implications on organization , especially on industrail relation.
Unions engage in political activities for different reasons .One such reason was that employee unions are attached to political parties for increasing their influencing power, (Sturmthal, 1972).Other reasons for unions to be politicized are , as a strategy for mobilizing more employees for their union actions and to balance the power in context where economic power is eroding (Piazza, 2001).Consisting with these propositions, Neeta, (2008), provided evidence of successful engagement of political activities by unions in wining workers" right.She claimed that unions need to keep a balance of their political engagement in the era of eroding the economic role of unions in developing countries in the face of economic liberalization and globalization.Some dimensions of union politicization have been identfied by researchers.Among indentified dimensions are political instrumentality ( Fiorito, 1987), political belief (Kelly, 1994;Charlwood, 2002), and employee voting intention (Leigh, 2006).However, political instrumentality and political belief are the two dimensions which have been attended mostly by researchers.

Political Instrumentality
Political instrumentality is the perception of union members that how far political activities of the union benefit them (Fiorito, 1987).However, it has been identified as a sub dimension of a broader construct known as union instrumentality.Union instrumentality is conceptualized as the beliefs that unions are able to improve the benefits for the members through collective bargaining and contact administration activities (Kochan, 1979).Early studies had identified two dimensions of union instrumentality as economic and non economic (Youngblood, William, & Angelo, 1982).Later, it was conceptualized as union leadership and external power (Craft & Suhail, 1983) This political instrumentality conceptually captures the political behaviour of union, measuring its politicality in oneway.
Though the political instrumentality is found to be explaining the political behaviour of unions as aggreagte unit, it is not explaining the political behaviour, or intention of union members at individual level.Later, reseachers identified another aspects of union politicization, explaining political belief of members literaly known as political belief or political ideology.

4.Political Belief of Union Members
Political belief or ideology is defined as the degree of perception of the members that union have to play a political role as well as the economic role in unionization process ( Kelly, 1994).Early record of identification of political view of employees as union dimension was recorded in the study of Adam, (1974).He claimed that the willingness of an employee to join with a union will be largely determined by his political view.Later, it was Kochan (1980) who distinguised between leftwing and right-wing ploitcal belief of union employees.Left-wing political belief is largely endosed with Marxist ideology where unions are considered as vehicle for class struggle (Neeta, 2008).
Therefore, unions should gain a greater span of control over workplace in order to aviod employee exploitation ( Barling, et al.,1991).On the other hand, right-wing political view says that it it through the demoractic and peaceful encounter that employee should win their rights in workplace (re).In examining the political view of employees, Charlwood (2002) and Boxall, Haynes, and Macky (2006) found that employees with left-wing political view will be more likely to join employee unions than employees with right-wing politcal veiw.Adding to this is the fact that left-wing politically motivated employee percieve managerial performance less favourably while right-wing employee percieve it more favourablly (Boxall, et al.,2006).
Given the fact that reasons and implications for union politics may vary across countries and time (Hyman & Gumbrell-McCormick, 2010), the relevance and the degree of important of union politicization dimension may differ from context to context.Therefore, an understanding of the context is needed before an empirical investigation is carried out.

5.Union Politicization in Sri Lanka
Employee unions in Sri Lanka specially unions in the public sector has been subjective to empirical investigation over the past years.It has been found that majority of public sector unions are politicized in term of affiation with the mainstream politics of the country ( Nanayakkara, 1988: Biyanwila, 2003).Kearney (1971) pointed out that the political activities of trade unions are partisan activities, and unions are primarily concerned with the winning of the political power of the party with which they are alined.All major political parties have their own trade unions considered as labour wing of their parties.As a result, employee unions have been an avenue for political parties to express their rivaltry to each other ( Jayawardena, 1972).
On the other hand, mojority of the unions are based on left-wing political ideology ( Uyangoda, 2000).However, politcal parties with right-wing ideology have their own trade unions in most of organizations.Therefore, it can be deduced that trade unions in Sri Lanka are affiliated with both left-wing and right-wing politcial ideology.
Political view of union members largely determine the nature of the activities a union engage in .For instance, Murrillo and Ronconi ( 2004) concluded that it was the political motivation of unions that make them more prone to strike behavior than other factors.
Based on the above empirical accounts, it is obvious that politicization is one of the major aspects of employee unions in Sri Lanka.Therefore, any examination of the union in general and union behavior in particular should include this union politicization dimension in Sri Lankan context.Measurement of union politicization is the major aspect of such empirical examination.The present study is aimed at testing the goodness of a measure of union politicization based on two identified dimensions of politicization, namely political instrumentality and political belief.

Politicization
Measurement of union politicization has to assess the two identified dimensions of politicization i.e. political instrumentality and political ideology.The measurement to be tested under this study was adopted from the 10 items union instrumentality measure of Fiorito (1987) and the 5 item measure of political belief of British Attitude Survey, taken from Charlwood (2002).
It was noted some overlap can be seen among researchers between items measuring political instrumentality and work place instrumentality.For example, some items of Kotchan"s (1979) big power dimension and Chacko and Green"s (1982) power dimension coincide with political instrumentality of Fiorito.But some items are not consistent.It may be due to the fact that both political instrumentality and work place instrumentality comes under a broader construct of union instrumentality and some of the items are difficult to distinguish whether measuring political instrumentality or workplace instrumentality ( For instance, Are unions more powerful than employer?).Therefore, it was decided to include items of both political instrumentality and workplace instrumentality in the measurement to be tested and extracted only the items that loaded high with political dimension at the initial testing.The items of the measurement were scaled with Likert"s five point scale from 1 anchored with strongly disagree to 5 anchored with strongly agree.

Sample
The sample of this study included 200 employees from the public sector organizations in Sri Lanka.Anyway, only 145 respondents responded to the distributed questionare of which 9 were found to be incomplete and were excluded, redusing the number of responses to 136 which is a 68% response rate.The was selected with stratified random sampling in order to ensure that appropriate number of employees are selected from sample organizations.
The profile of the sample is such that 64.7% were male and 35.

Data Analysis
Exploratory factor analysis (EFA) and reliability analysis were used to test the validity and the reliability of the measurement of the three employee outcomes.It was intended to use exploratory factor analysis on the ground that no previous test of these measures have been done in Sri Lankan context.Byrne,( 2010) claimed that EFA is designed for situation where links between the observed and latent variables are unkonwn or uncertain.It will help researcher to determine how and to what extent observed variable are linked with the underline factor.Other researchers too have used EFA for the testing of goodness of measures of various variables (Wlliams & Anderson, 1991).
Cronbach Alpha is widely used relaibility coefficeint for assessing the reliability of a measurement instrument.It measures the degree to which the test score indicates the status of an individual item on the factor defined by the test , as well as the degree to which the test score demostrate individual differences in these traits (Cronbach & Meehl, 1955).Cronbach Alpha specifies that if the sum of the individual item variance is closer to the variance of the entire scale, alpha value come closer to zero, stating that they are not measuring the same construct.On the other hand, if the variance of the entire scale is much larger than the sum of the variance of individual items, alpha get a value closer to one.

The Result
An initial factor analysis was carried out with the 15 items with the aim of identifying the factor structure of the measurement instrument of which the result is shown in the following table Table 1 Rotated Component Matrix   Items  Component   1  2  3 1.Unions give members a service worth for the member fee they pay.

Unions have lot of influence over
what laws are passed in the country.
3. Unions have lot to say about who get elected to the power in the country.
4.Unions are more powerful than employers in the country 5. Unions have lot to say in how the country runs.
6.There is one law for employees and another law for management in this organization.
7. There should be a fair redistribution of rewards between the management and employee in this organization.
8. Business benefits owners at the expense of employees.9. Employees do not get a fair share of the contribution they make in the organization. .800 .

719
.702 .687 10. Management always tries to get the better of the employee.
11. Union requires members to go along with decision they don"t like.
12. Unions improve the job security of the workers 13.Unions improve the wages and working conditions of workers.As shown in the table I, the initial factor analysis with all items of workplace instrumentality, political instrumentality and political belief extracted three factors measurement model.However, one item of the workplace instrumentality (unions give members a service worth for the member fee they pay) was loaded with the political instrumentality (Factor I) and the remaining items of workplace instrumentality were loaded with that factor (Factor3).The items of political belief dimension were loaded with each other as initially expected (Factor 2).
Since the very objective of this study is to test the goodness of the union politicization measurement adapted, factor analysis was done again by excluding the workplace instrumentality items and the result is given in the table II 1. Unions give members a service worth for the member fee they pay.
2. Unions have lot to say about who get elected to the power in the country.
3. Unions are more powerful than employers in the country 4. Unions have lot to say in how the country runs.
5. Unions have lot of influence over what laws are passed in the country.
6.There is one law for employees and another law for management in this organization.
7. There should be a fair redistribution of rewards between the management and employee in this organization.
8. Business benefits owners at the expense of employees.9. Employees do not get a fair share of the contribution they make in the organization.
10. Management always tries to get the better of the employee. .

Discussion and Conclusion
The objective of the study reported here was to test the goodness (Validity and Reliability) of measure of union politicization adapted from political instrumentality of Fiorito (1987) and measure of political belief of Charlwood (2002) in Sri Lankan Context.The measure tested incorporated both political instrumentality and political belief dimensions given the fact that unions in Sri Lanka, particularly unions in the public sector are highly politicized (Biyanwilla, 2003) and motivated by political ideology ( (Uyangoda, 2000).
The initial Exploratory Factor analysis with principle component analysis and varimax rotation, extracted a three factor model of measurement with work place instrumentality, political instrumentality and political belief dimension.A noteworthy observation in the initial analysis is that, one of the items for workplace instrumentality (unions give members a service worth for the member fee they pay) was loaded high with political instrumentality dimension.This may be due to the fact that respondents might have perceived that most of the benefits they received from unions are due to political activities of the union they represent.
However, the second analysis, by excluding items of workplace instrumentality, it was extracted two factor model of measurement loading each items with respective dimension.
The first factor identified as political instrumentality recorded the highest value for percentage of variance explained (33.7%).Political belief dimension explain only 21.7% of variance in political behavior.Therefore, it can be inferred that political instrumentality is the most significant factor for explaining political behavior of unions in Sri Lanka.This is in line with the findings of other studies that union members perceived that workplace benefits can be improved by unions engaging in political activities.It is because union can influence the policy decisions of management largely with political behavior ( (Jinadasa & Opatha, 1999).
The reliability level of each dimension recorded high (Chronbach Alpha=.82,.75)indicating a high reliable level.Further, since the correlation between the two dimensions was moderate level, the descriminant validity of the measure can be assumed to be high.Two dimensions identified; political instrumentality and political belief seem to be highly relevant in explaining union politicization in Sri Lanka.Therefore, the measurement tested by this study can be used validly and reliably for measuring the union politicization in Sri Lankan context.