Chapter 2 Literature Review
2.3. Service Quality Perceptions in Call Center
2.3.4. The Role of Language in Call Center Context
2.3.4.4. Common Differences Between Language Spoken by Local and
It has been recommended that second language acquisition is relatively complicated process with many factors influencing the final output. It is, however, debatable to what extent the output is final since language studying is in many cases life lasting process. It has been proved that local speakers can recognize a foreign accent from the very short speech samples (Flege and Hammond, 1982; Flege, 1984 in Magen, 1998).
In the following subchapters some of typical features differing non-local from local speakers are presented.
The forthcoming aspects form a basis for the thesis´s empirical study. Since one of the thesis aims is to identify language factors contributing to the perceived quality, the following aspects are important part of empirical research.
28
1. Accent
The second language acquisition has been gaining the research interest within the field of psychology and social psychology for decades (Derwing, 2003). Accent as the important aspect of the second language use has been given attention as well. Topical research has focused on the communication from both parties, local speaker´s (Magen, 1998; Cargile, 2000) and non-local speaker´s (Derwing, 2003; Gluszek and Dovidio, 2010). Sociolinguistics literature (Lippi-Green, 1994; Giles and Powesland, 1975) considers accent as important indicator of person’s ethnicity, regional association and social background. This fact suggests that non-local speakers are assigned certain characteristics formed apart of their appearance by the way they speak the language.
In personal encounters the total impression is created by both physical appearance and spoken language together with non-verbal signals.
As noted above, customers can recognize the non-local speakers already after a couple of exchanged sentences. Depending on pre existing attitudes, established and enforced throughout an individual’s life experiences clients evaluate AGENT and the company (Barker and Härtel, 2004). As accents are often connected with certain status (e.g. social and cultural background) people tend to react differently on various accents (Magen, 1998; Cargile, 2000). As a consequence some non-local speakers are perceived more positively than others. The accent indicates in many cases the non-local speaker´s probable country of origin. Some researchers (Thelen et al., 2010) suggest that this association of Customer Care being located in a particular country significantly influences the perceived service quality. Chapter 2.3.2.2. Country Image further examines this concept.
29
Wang et al. (2009) have conducted a study where respondents listened to the short conversation samples from call centers. The conversation content was indistinguishable in all cases; manipulation factor was Customer Care employee´s accent. On the basis of the research Wang et al. (2009) conclude that customer satisfaction is strongly affected by the service outcome. In case that clients are not satisfied they tend to reflect this fact on the employee personality. Thus, Indian English has received the least positive feedback out of offered options (Wang et al., 2009).
Nevertheless, the researches mentioned above were carried out in the United States of America and it is debatable whether their findings relate also in Asian setting differing from Anglo-Saxon culturally.
2. Vocabulary
Customer Care representative needs to master language to such extent that
communication with client makes no problems to any side. A sufficient vocabulary is a core condition for a successful interaction. However, as far as it concerns the second language acquisition, there may appear challenges related to vocabulary. Typically, non-local speakers face four kinds of problems:
30
(1) Insufficient vocabulary
It is very difficult to determine the proficiency level which can be considered as a sufficient. Generally, Customer Care employee has to be familiar with the second language vocabulary at least to such extent that enables understanding customer´s requirements and providing an understandable good-proficiency answer in a fluent matter. Naturally, the vocabulary sufficiency depends on the kind of Customer Care.
Vocabulary proficiency requirements are different for the employee accepting and confirming simple order by phone and employee advising customer how to set up a computer program.
(2) Use of words in a wrong context
This problem is strongly related to language incompetence. Employee has learnt to use a word in a wrong context changing the sentence meaning. Consequently, this influences the customer´s image of employee and the company in a negative manner.
(3) Unfamiliarity with idioms
To understand idioms, usually specific in every language; non-local speaker has to be on the advanced language level. Various languages express idioms in various ways often leaving non-local speakers confused about the meaning.
(4) Vocabulary confusion
When using the second language, people are likely to bring cognitive frames from mother tongue and so ultimately use improper wording (Marcella and Davies, 2004).
Commonly these mistakes include word confusion such as using a word from a one language in another language. Additionally, non-local speakers may not be aware of words possible other meanings.
31
3. Syntax
Syntax is defined as the study of the principles and processes by which sentences are constructed in particular languages (Chomsky, 1971). Naturally, this process varies from language to language. In some of them free word order applies while in others fixed word order needs to be kept. The basic rules of second language syntax need to be studied by non-local speakers already at the very beginning of second language acquisition. The knowledge of syntax is crucial for a sentence forming.
For instance, question form in some languages is formed by switching the pronoun and verb position in the sentence. Leaving this out, a person does not ask question but instead states a sentence. Alter locally, especially in free word order languages manipulating with a word order changes noticeably the sentence meaning. Non-local speakers may not be aware of the shifted sentence meaning.
4. Rhythm of speech
Rhythm of speech significantly influences the understandability of message.
Therefore, both client and agent should adjust the speech pace so that interaction makes no problem to any of them. However, if they communicated grammatically correctly and their speeches are otherwise understandable, listeners are not very sensitive to voicing differences (Magen, 1998).
32
2.3.4.5 Language factors influencing the service quality perception
Chapter 2.3.4.4. Common differences between language spoken by local and non-local speakers identified the main factors distinguishing non-local and non-non-local speech.
The figure 2 summarizes the language factors contributing to the call center interaction between customer and agent.
Figure 2 Aspects affecting the call center service interaction with non-local speaker (Flege and Hammond, 1982; Flege, 1984 in Magen, 1998)
Even though the above mentioned factors are affecting the service interaction, their contribution does not need to be equal. For instance, the accent may be perceived more significantly than rhythm of speech by customer (local speaker). The perception of these factors depends on the concrete customer and concrete service encounter episode.
Service Interaction
Accent
Vocabulary
Syntax Rhythm of
Speech Cognitive Language Frame
33
SERVQUAL and RECOVSTAT measurement scales were discussed in the section 2.3.3. Service quality perception measurement. Both tools were used in the call center settings and therefore, this study as well, takes advantage from them.
2.3.5 Summary
Theoretical part centered on the three major areas, the call center essentials, service quality perception and image in the call center context and the role of language in the call center context.
Studies indicate that countries are assigned a certain image in the client´s eyes and this fact also influences on service quality (Huber and McGann, 1982). Therefore, the company and country image aspects were discussed as vital pointers of customer´s service evaluation.
Due to thesis focus, attention was given to the role of language in service interactions.
The language is the very vital part of any interaction, especially in call centers where the other visual components are missing. The language defined as a fence created by the circumstances when customer and agent do not share the same mother tongue (Stringfellow et al. 2008) can make the interaction more challenging. For offshored call centers the significant percentage of non-local speakers is typical. The situation when client and agent do not come from the same country spaced out from language also cultural and geographical distance related barriers (Stringfellow et al., 2008).
Since the language is a significant part of local uniqueness (Isaacs, 1975; Sonntag, 2009), the role of language as a local aspect was discussed. As Holmqvist (2009) states out many people love their language that represents their identity and being
34
affected by their own cultural as consumers. Speaking the language is by some people perceived as a belonging to the nation.
Due to thesis theme the second language acquisition was given the attention. Ongoing research aims for determining whether a non-local speaker is capable to master the language at the local level (Robert and Miyake, 2006). Due to the fact that local speakers recognize non-local speech in a short time (Flege and Hammond, 1982;
Flege, 1984 all cited in Magen, 1998), the subsequent chapters discussed in detail common characteristics of non-local speech: accent, vocabulary, syntax, rhythm of speech and different cognitive language frame.
35
2.4 Hypotheses
Based on the literature review, five hypotheses were recommended. H1 and H2 were formed based on the findings related to offshoring awareness and call center allocation abroad and company´s perceived trustworthiness respectively. H3 is directly connected to the thesis aim; exploring the influence of non-local speech on the perceived quality. H4 and H5 support H3 trying to further explain the role of accent (H5) in the interaction and non-local speakers perception (H4).
H1: Interacting with non-local speaker, customers are more likely to suppose that call center is located outside of country
H2: Customers perceive non-local Customer Care representative as less trustworthy compared to local speaker
H3: Customers perceive lower service quality when interacting with non-local Customer Care representative
H4: Non-local Customer Care representative is perceived as less competent
H5: Customers perceive higher probability of misunderstanding when interacting with non-local speakers
36
Chapter 3 Methods
This chapter presents the general idea of selected research methods and explains the incentive behind these choices. The need for pilot study is justified, with the main aspects of the chosen survey approach of the empirical study.
3.1 Selection of research method
The intention of this thesis is to determine whether clients perceive communication differently if service is provided by non-local agent. Moreover, this thesis also plans for determining the part of respondents‟ age in the service quality perception. The selection of research method is very significant decision in regard to thesis. When choosing the most appropriate research method, the thesis seek, especially the research question, need to be taken to account. At first, researcher has to make a decision whether to use qualitative or survey methods or eventually their mixture.
Even as survey methods deal with impartially measurable data, qualitative focus on deep analysis of issue. In marketing, the survey methods used to be chosen to qualitative since they base on the statistics and are therefore easily provable (Hunt, 1994). Qualitative methods intend in prior for understanding a certain phenomenon than to simplify the truth (Patton 2002). While Reichardt and Cook (1979) put it, the primary difference between survey and qualitative methods is in the measurement of verification versus discovery. Survey techniques are frequently used to prove or validate theories whilst qualitative methods are used in theory creating and developing.
37
3.2 Research design
This part deals with the research design of the pilot study and survey. Regarding the first mentioned, the methods used for the interviews are justified and the interview guides are offered. In the survey part, source of data and questionnaire and general research design are explained.
3.2.1 Pilot study
The pilot study was derived from the interviews with two collections of informants, clients and Customer Care agents.
3.2.1.1 Participants of pilot study
10 respondents (5 answered the local speaker version and 5 respondents filled the non-local speaker version questionnaire) were conducted in order to determine possible unfairness. To reduce error it was aimed for made scenario to be realistic and clear.
Customer
Participant Age Nationality Language
A 45 Taiwan Chinese
B 51 Taiwan Chinese
C 24 Taiwan Chinese
D 28 Taiwan Chinese
E 34 Taiwan Chinese
F 28 Philippines Tagalog/English
G 35 Philippines Tagalog/English
H 26 Philippines Tagalog/English
I 28 Philippines Tagalog/English
J 62 Philippines Tagalog/English
38
Customer Care Representatives:
Participant Nationality Language Language Customer Care Provided in
A Taiwan Chinese Chinese
B Taiwan Chinese Chinese
C Taiwan Chinese Chinese/English
D Taiwan Chinese Chinese/English
E Philippines Tagalog/English Tagalog/English F Philippines Tagalog/English Tagalog/English G Philippines Tagalog/English Tagalog/English
3.2.1.1 Interviews with clients
When taking qualitative approach interview is a literally standard tool to obtain information. They are interviewed by researcher in order to get information that is unobservable, like feelings, thoughts and intentions (Patton, 2002). During interviews researcher has chance to discover a person´s behavior and value (Silverman, 2006).
However, interviews are normally not practiced in the survey methods mostly due to time and financial demand.
The thesis centers on exploring the power of language on the perceived service quality in the context of Asian countries and languages with emphasizes on the Chinese and Tagalog language. Due to the thesis being conducted in two countries, ten interviews with customers were carried out with five respondents of Taiwanese nationality and five of Filipino nationality. The major reason for carrying out the interviews with clients was to gain a pre-understanding on the topic prior to the survey study and then search for likeness and dissimilarities with non-Asian research.
39
Hypotheses were recommended after the pilot study findings.
Interviews with Taiwanese informants were conducted during March, 2014 in Taipei, Taiwan. Five face-to-face interviews lasted about 10 minutes each. Likewise, five interviews with Filipinos were carried out with around same duration but interviewed conducted via Skype. These interviews took a place in March/2014 in Taipei.
Several methods may be taken in regard to case selection. For example, Patton (2006) varies among sixteen different techniques within two categories – random probability and purposeful sampling. In qualitative interviews the number of informants is rather small with the study aim being to obtain in-depth insight to a problem. Therefore, as Eisenhardt (1989) stated, cases selection is vital for creating theory. In this study the purposeful sampling was used leaving on the researcher to choose the cases for own research purposes.
Informants were selected in regard to the purposive sampling, where it is the researcher choosing respondents on the beginning of their suitability with the study aim. As Patton (2002) states it the information-rich cases are to be selected. Therefore, within the purposive sampling a criteria selection has been applied. The informants had to complete the condition of at least one communication with a call center prior to interviews.
40
3.2.1.2 Interview guide for clients
Interview guide is used when researcher performed the interview guide method when the interview proceeds on the basis of a set of predefined questions. In this case, researcher may follow exciting way in conversation returning later back to the guide process. Aside from the open ended questions, they were described a short scenario of calling to a call center and then asked to answer some questions. The incentive for including this short scenario was for researcher to get a real picture of their attitude towards one specific situation.
3.2.1.3 Interviews with Customer Care agents
The interviews with the agents were significant part of this these since these employees, all non-local speakers, communicate or communicated with clients – local speakers on the daily basis. In this situation, the criteria sampling was used as well and prior to the interview informants had to have a work experience as Customer Care agent in a call center. It was very valuable to get an insight from this and the interviews were source of great information on the non-local speakers. Agents were chosen to represent maximum variety of Asian languages and countries. The interviews were conducted during March 2014 in Taipei. Seven face-to-face interviews lasted approximately 20 minutes each.
3.2.1.4 Interview guide for Customer Care agents
The interview guide for agents includes to a high extent similar questions as the interview guide for clients. However, there are some differences too. Firstly, it takes
41
the agent´s opinion and it is based rather than on scenario on individual experiences.
Additionally the background questions are expanded by the set of enquires concerned with the second language acquisition.
3.2.1.5 Questions design for interview
As stated above, the interviews with clients and agents formed the base for determining the aspects used in the survey part. Thus, categories found by the pilot study analysis were developed when designing a questionnaire.
Questionnaire has been chosen as the most suitable method for the data gaining.
Questionnaire has a multiple use often being utilized in experiments and case studies.
Due to the fact that one of the study purpose was to explore whether perception of non-local speakers differs between two countries, Taiwan and Philippines, the questionnaire was distributed in two geographically different locations and moreover in two language versions.
Scenario experiment was used as the most suitable choice for the study aim. Saunders et al. (2009) stated that experiments are considered to explore whether there is a relationship between two variables. In experiments respondents are normally assigned into two groups, experimental and control. Assigning to groups occurs on the random basis. In this thesis, respondents receiving “non-local speaker” version of scenario are members of experimental group whilst those getting “local speaker” version are considered as a control group. Roggeeveen et al. (2007) used in the similar study the scenario based questionnaire with the own research scale items developed. As they point out scenario based technique allows controlling the independent variables and guarantee that all respondents are given the same treatment. The classical form of
42
questionnaire with questions and answers based on the respondent´s knowledge did not go well with this study aims since prior to the survey study a comprehensive inquiry would have to be carried out to find only respondents having a prior experience with the non-local speakers in call centers. Furthermore, the types of call centers would most likely significantly vary making experiences of respondents non-comparable. Thus, the scenario experiment when respondent is presented with a story and asked to imagine being the part of it seems to be the most suitable selection. Once scenario was presented, a set of follow up questions was introduced after the play. The scenario was printed in two versions distinguishing only in the characteristic of call center employee being and not being the language´s local speaker. The presented scenario (purchasing a membership and later making complaint in Customer Care) received a positive feedback during the pilot study with respondents stating that scenario was easy to imagine and represented a type of situation most of the people have already experienced. The fact that all respondents base their answers on the same text can be seen as the advantage of Scenario Experiment. However, the scenario is only a made-up and respondent´s answer does not need to necessarily reflect their real behavior when it comes to the situation.
3.2.2 Survey study 3.2.2.1 Samples
In the survey section, a total 280 questionnaires were submitted and a total of 241 questionnaires returned. Nevertheless, respondents marking “Other” as their mother tongue were excluded due to thesis purpose being narrowed to comparison between Filipinos and Taiwanese. Thus, after exclusion of 15 questionnaires, the sample consisted of 226 respondents, 91 (40.3%) males, 135 (59.7%) females. The local
43
language being used is Taiwanese (114, 50.4%) and Tagalog (112, 49.6%), respectively. The age groups of respondents are drawn in the figure 3.
Figure 3. The average age of participants
Figure 3. The average age of participants