Actual purchase behavior of edible bird’s nest products in malaysia using cluster analysis

Actual Purchase Behavior of Edible Bird ’s Nest Products in Malaysia Using Cluster Analysis Nurul Nabilah Huda Mohamad Shukri, Nolila Mohd Nawi, Amin Mahir Abdullah, and Norsida Man Depa

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=wfpm20

ISSN: 1045-4446 (Print) 1540-4102 (Online) Journal homepage: https://www.tandfonline.com/loi/wfpm20

Actual Purchase Behavior of Edible Bird’s Nest Products in Malaysia Using Cluster Analysis

Nurul Nabilah Huda Mohamad Shukri, Nolila Mohd Nawi, Amin Mahir Abdullah & Norsida Man

To cite this article: Nurul Nabilah Huda Mohamad Shukri, Nolila Mohd Nawi, Amin Mahir

Abdullah & Norsida Man (2019) Actual Purchase Behavior of Edible Bird’s Nest Products in Malaysia Using Cluster Analysis, Journal of Food Products Marketing, 25:8, 849-860, DOI: 10.1080/10454446.2019.1691105

To link to this article: https://doi.org/10.1080/10454446.2019.1691105

Published online: 12 Nov 2019.

Submit your article to this journal

Article views: 24

View related articles

View Crossmark data

Actual Purchase Behavior of Edible Bird ’s Nest Products in Malaysia Using Cluster Analysis

Nurul Nabilah Huda Mohamad Shukri, Nolila Mohd Nawi, Amin Mahir Abdullah, and Norsida Man

Department of Agribusiness and Bioresource Economics, Faculty of Agriculture, Universiti Putra Malaysia, Serdang, Selangor, Malaysia

ABSTRACT

The focus of this study was to classify actual buying behavior

of edible bird ’s nest (EBN) products into relatively homogenous

groups for the purpose of market segmentation Determining

the profile of actual customers could provide direct

character-ization of products that should be displayed on the shelf The

data were collected by quota sampling with 312 buyers

parti-cipated in the survey Three types of cluster models (K-means,

Kohonen, and two-step cluster) were estimated and compared

with in order to identify an optimum number of clusters and

capture group with a similar pattern Foremost, a two-step

method was chosen for analysis as it provides better average

silhouette and well-defined of group segments than the other

two cluster methods The results provide specific

characteriza-tion of buyers that will be useful to build a target list of

potential customers and develop new offerings to meet the

new needs of customers for EBN products in Malaysia.

KEYWORDS

Consumers ’ profile; food safety; health conscious; market segmentation; spending behavior

Introduction

Ideally, the government especially the Ministry of Agriculture (MOA) would want to protect the local edible bird’s nest (EBN) industry and raise demand for EBN There are two divisions under Department of Veterinary Services (DVS), which are Downstream Industry Development Division (DIDD) and Livestock and Commodity Development Division (LCDD) that provide assistance for swiftlets’ ranchers to further sustain high quality of local products in the market The functions of DVS are to (i) control, prevent, and eradicate animal disease and zoonotic; (ii) livestock production and animal feeds; (iii) inspect ranching houses and processing plant; (iv) control import and export of livestock; (v) provide training for industry livestock; (vi) product development and animal health as well as public health; (vii) research on animals and genetic diseases; and (viii) control animal welfare and genetic materials On top of that, they also provide financial incentives to

CONTACT Nolila Mohd Nawi nolila@upm.edu.my Department of Agribusiness Bioresource Economics, Faculty of Agriculture, Universiti Putra Malaysia, Serdang, Selangor 43400, Malaysia

https://doi.org/10.1080/10454446.2019.1691105

swiftlets’ ranchers for product development and market positioning According to Z Hussin (Personal communication, August 8, 2018) from LCDD, Malaysia is a country with the least import of EBN than other countries This is due to the fact that Malaysia is a producer of EBN and has reported as the third exporter after Indonesia and Thailand (Kuan & Lee,

2005; Trade statistic for international business development,2018) Over and above, Dr Quaza Nizamuddin, the deputy director from DVS, claimed that Malaysia has produced 350 metrics tons of EBN in 2016 and around 150 metrics tons of EBN import was expected in the same year (Tan,2016) And

in the recent year, about RM 200 million of EBN has produced only at Sarawak state (“Bird’s nest raked in 200mil”, 2018) and exported value of EBN from Malaysia has reached to RM 1.32 billion (Chern,2019) In order to expand the product range of EBN, the Malaysia government has allocated

RM 8.2 million for research funding (Tey, 2015)

different countries and the properties are slightly different Researchers have listed five main elements found in EBN, such as protein, carbohydrate, moisture, ash, and lipid Marcone (2005) identified Indonesia’s EBN com-posed with 62% of protein, 27.26% of carbohydrate, 7.5% of moisture, 2.1%

of ash, and 0.14% of lipid Meanwhile, Saengkrajang, Matan, and Matan (2013) attested that the contents are varied for protein (60.9% to 66.9%), carbohydrates (25.4% to 30.7%), moisture (17.81% to 24.3%), ash (5.9% to 7.4%), and lipid (0.4% to 1.3%) Further, Hamzah, Hulwani Ibrahim, Hussin, Hashim, and Lee (2013) and Norhayati, Azman, and Nazaimoon (2010) have recorded the contents found in Malaysia EBN as shown inTable 1 Previous study has revealed the nutritional contents of EBN, and this evidence pro-vides support that EBN from Malaysia are of good quality to compete with EBN from other countries

The issues with contamination and adulteration in EBN products in the market are threatening the Swiftlet ranching industry in Malaysia Contamination can be defined as unwanted microbial in foods that can cause foodborne disease Thereby, contaminated food appears to be grow-ing concern which can cause serious inferiority to health (Adeola, 2000; Costa, Giordano, Tagliaferri, Caglieri, & Mutti, 2009; Magalhães, De,

Table 1 Comparison of nutrition contents.

Indonesia (Marcone, 2005 )

Thailand (Saengkrajang

et al., 2013 )

Malaysia (Hamzah et al., 2013 ; Norhayati

et al., 2010 )

Sources: Mohamad Shukri et al ( 2018 )

Soares, & Azevedo, 2001; Reilly, 2002; Zhuang, Zou, Li, & Li, 2009) The most serious contamination in EBN was nitrate content found above permissible level of 34 ppm in 2011, causing the price to fall nearly 50% from its original price (Cheng,2016) The recent avian flu (H5N1) strikes

in Kelantan (Mohd Shahar, 2017) have prompted the same temporary ban from China back in 2011 and then in 2017 Contamination in EBN appears

to be growing concern and imposed threats on the safe consumption of this product In addition, adulteration in food has been theorized as

a similar kinship or feature with problematic ingredients that can cause serious adverse health effects (Hohl & Gaskell, 2008) Hence, adulteration

in EBN poses vulnerable risks to the public health This has resulted in anxiety to the public of the consumption of the highly purported nutritious products In order to regain the public’s trust and confidence, the Malaysian government has made big efforts resulting in a steady boost of the local bird’s nest price Previous studies were mainly emphasizing on the biological aspects of the products Unfortunately, the public’s percep-tion and behavior in purchasing local EBN after the issues have been resolved are yet to be determined Rising concerns in health and food safety are changing customers spending behavior that influence their purchasing decision In regard to the current situation, there is little information on the customers’ behavior and their concern in consuming EBN Therefore, the focus of this study was to classify actual buying behavior of EBN products into groups that are relatively homogenous for the purpose of market segmentation Specifically, the study objectives are

to determine (i) the best cluster method in segmenting buying behavior for EBN product, (ii) the profile of EBN buyers, and (iii) the level of health conscious and food safety concern among EBN buyers Understanding the customers’ spending behavior is important to any organization for strate-gically targeting the right customers Hence, determining the profile of actual customers could provide direct characterization of products that should be displayed on the shelf

Research methodology

Respondents were queried to respond to questionnaires related to their health-conscious level and safety concern on EBN products The ques-tionnaire was adapted from Muhammad, Sherif, and Gheblawi (2010), Kraft and Goodell (1993), Darian and Tucci (2013), and Wee et al (2014) The structured questionnaires include five points Likert scale for respondents to rate their level of concern from 1 (strongly disagree) to 5 (strongly agree) Respondents were also required to complete a set of demographic questions for the purpose of segmentation The data were collected by quota sampling (through ethnic-based population: 61.7% for

Malays, 20.8% for Chinese, 6.2% for Indian, and 0.9% for other ethnics), and the questionnaire was distributed at shopping malls around the Klang Valley area About 1,000 of respondents who live in the Klang Valley participated in the survey

Data cleaning was conducted to ensure the quality of data fulfilled the analysis criteria The first assessment was to determine participants who actually purchase EBN while the remaining of non-buyers were excluded from the dataset The second assessment was to identify the missing cases and outliers and then removed from the dataset resulting in 312 respon-dents who were actually purchasing EBN from 1,000 participants In this study, two types of data analysis were performed to analyze the data, namely descriptive analysis and cluster analysis using 312 samples of buyers The dataset was converted into categorical (nominal and binary) variables for the analysis Descriptive analysis is a statistical analysis used

to summarize the socio-demographic characteristics of the samples using frequency and percentage Cluster analysis, on the other hand, is

a statistical method used for segmentation techniques to explore salient features and the degree of association between possible numbers of cluster for the given data The cluster technique does not require a distinction between independent and dependent variables Hence, all variables were standardized and assigned as input and no target field required This is performed to evaluate the possibility that achieved based on homogeneity

or similarity in customers’ spending behavior which falls in the same group The cluster analysis was incorporated using the SPSS Modeler software 18 and several cluster models were generated and compared with Auto Cluster node Three cluster models (K-means, Kohonen, and two-step cluster) were estimated and compared to identify the optimum number of clusters and capture group with a similar pattern K-mean cluster is the most popular clustering technique because of its easily programmed and computationally economic (MacQueen, 1967) whereby the n-dimension of sample is split into K clusters The idea of partition

in K-means cluster is to minimize the sum of squared error to compute possible clusters Meanwhile, Kohonen network is a cluster technique that has artificial neural network learning criteria invented by Teuvo Kohonen

It is a faster algorithm of visual clustering techniques that enable descrip-tion of clusters optimally for segmenting the same pattern as the cluster network forming two-dimensional grid (Kohonen, 1998) This type of cluster is also known as self-organizing map algorithm which consists of two connected layers of inputs and output (cluster) No partition is gen-erated but the minimum weight distance establishes the same pattern For two-step, the methods provide and describe the interrelationship between variables and segment profiles (Satish & Bharadhwaj, 2010) The method was established by Chiu, Fang, Chen, Wang, and Jeris (2001)

Results and discussion

The socio-demographic profile of buyers is presented as categories in

Table 2 for all the characteristics Female (72.4%) were more likely to buy EBN products compared to male (27.6%) Malays contributed 75.3% in purchasing EBN products granted that Malays as the majority of the ethnic group in Malaysia whereas the remaining 24.7% attributed by other groups

of ethnic (Chinese, Indian, Aborigines, Sabahans, and Sarawakians) Islam, being the official religion in Malaysia, constituted 76.3% of the total buyers and the remaining 23.7% consisted of other religions (Buddhist, Christians and Hindu) Respondents with tertiary level of education (77.6%) were the predominant buyers More than half of the buyers were married (57.4%), followed by single individuals (41.0%) and divorced (1.6%) There was

a slight difference among the proportion for age group from 18 to

27 years old (31.4%) and 28 to 37 years old (34.0%) Older age group contributed small percentage of proportion compared to the prior age group, 38 to 47 years old (21.2%), and above 48 years old (13.5%) Almost half of the buyers earned a monthly income between RM 2,001 and RM 4,000 (38.1%), and more than half of the buyers worked in private sec-tor (57.4%)

The Auto Cluster node was performed, and the three cluster methods were compared The finding indicates that the number of clusters found for a two-step was two clusters, for K-mean was five clusters, and Kohonen was 12 clusters The average silhouette for a two-step, K-mean, and Kohonen was 0.4 (fair), 0.3 (poor), and 0.2 (poor), respectively Hence, a two-step method was chosen for analysis as it provides better average silhouette, optimum number

of clusters, and well-defined group segmentation than the other two cluster methods

Table 2 Socio-demographic profile of buyers.

Foremost, two distinct groups (segments) from 12 variables were found using a two-step method and the results are delineated in Table 3 The first cluster with the total of buyers 235 shows that 100.0% were Malay and they embraced Islam as their religion (100%) About 49.1% of buyers worked in the private sector and they earned an income less than RM 3,500 (68.4%) per month Besides, they were influenced either by friends, advertisement or relatives to purchase EBN products with the percentages of 25.6%, 18.4%, and 10.7%, respectively The buyers in this cluster were below 40 years old and they have tertiary education (79.9%) Most of them purchased EBN products either at the drug store or pharmacy (52.6%), health center (27.4%) or online store (11.1%), and spent less than RM 100 per month Buyers in this cluster were very concerned about their health level (90.2%) and food safety (94.5%)

Table 3 Two-step cluster analysis results.

The opposite pattern was observed in the second cluster as the majority were non-Malay (100%) and embraced different religions than Islam (Christian, Buddha, and Hindu) There were not much of differences corre-sponding to the first cluster where they were about 40 years old had tertiary education (70.1%) and worked in private sector (83.1%) Buyers in cluster 2 also preferred to purchase EBN products either at the drug store or pharmacy (50.6%), health center (39.0%), or online store (6.5%) However, non-Malays seemed to have a better monthly salary (more than RM 3,501) and they invested slightly more on EBN products than Malays as the amount spent was over RM 100 per month The main influencers to buyers in cluster 2 were parent (36.4%), relatives (15.6%), friends (15.6%), and coworker at the workplace (11.7%) Therefore, there was no apparent contradiction to cluster

1 Comparable percentages were found on their level of concerns in health (94.8%) and food safety (90.9%) for non-Malays The distinct differences between the two clusters showed that similarity within groups form from the same features characteristics and dissimilarity in different groups (Pride, Ferrell, Lukas, Schembri, & Niininen, 2012)

Several questions were solicited regarding their consciousness in health that have influenced them to purchase EBN products and the results are shown in Table 4 In review, majority of the respondents agreed with the statements About 46.2% agreed that “I will buy bird’s nest products, since

I am very self-conscious about my health,” while 1.3% strongly disagreed Furthermore, 43.9% of buyers willing to spend on bird’s nest products “I am willing to invest some money to buy bird’s nest products for my health” and only 1.6% strongly disagreed The buyers were skeptical with the contents in bird’s nest and they were either agreed or strongly agreed on “I will consume bird’s nest products, if the products have not contain any unhealthy sub-stances.” Since they were aware in the health benefits of bird’s nest products, 44.2% agreed that “I intend to buy bird’s nest products for my long-term health benefits” and none has disagreed with this statement Similar with

“bird’s nest products have high nutritional value which good for my health,” 46.5% of buyers agreed with the statement

Table 4 The frequency level of health conscious.

ITEMS

Strongly Disagree Disagree

Moderately

Strongly Agree

I will buy bird ’s nest products, since I am very

self-conscious about my health

1.3%

(4)

3.8%

(12)

21.2%

(66)

46.2%

(144)

27.6% (86)

I am willing to invest some money to buy bird ’s nest

products for my health

1.6%

(5)

3.8%

(12)

24.0%

(75)

43.9%

(137)

26.6% (83)

I will consume bird ’s nest products, if the product

has not contain any unhealthy substances

0.0%

(0)

1.9%

(6)

13.8%

(43)

42.6%

(133)

41.7% (130)

I intend to buy bird ’s nest products for my long term

health benefits

0.0%

(0)

2.2%

(7)

19.6%

(61)

44.2%

(138)

34.0% (106) Bird ’s nest products have high nutritional value

which good for my health

0.3%

(1)

3.5%

(11)

17.6%

(55)

46.5%

(145) 32.1% (100)

Food safety is an important aspect in purchasing bird’s nest product as recent news breakdown the world with high nitrate level found in bird’s nest can deteriorate the consumers’ health (Mohamad Shukri, Nawi, Abdullah, & Man, 2018) A total of 236 buyers stated that it is extremely important for bird’s nest to be “100% genuine and safe to consume” as shown in Table 5 They also agreed that it is extremely important (71.5%) for“durability on the label is provided.” Thirdly, “visually display actual product contents” was extremely important with 61.5% of buyers agreed on it and followed by

“quality logo on the label is provided” (59.9%), “the nutrient information

on the label is provided” (55.4%), “no preservative in EBN products” (44.9%), and“safe from hazardous substance such as nitrate” (44.6%) However, both country of origin and country of manufacture were important information for the buyers with 44.6% and 43.3%, respectively

Conclusion and implications

In a nutshell, two-step cluster provided optimal number of clusters than other methods The findings showed two groups of clusters were iden-tified from 12 variables for EBNs’ buyers, categorized as low spender and high spender Cluster 1 comprised of low spenders who spent less than RM 100 on EBN products as their monthly salary below RM 3,500 The members of this cluster were Malays and they were influenced or encouraged largely by their friends to purchase EBN Cluster 2, on the other hand, comprised of high spenders who spent more than RM 100

Table 5 The frequency level of food safety among buyers.

ITEMS

Not at all important

Slightly important

Somewhat

Extremely important Safe from hazardous substance

such as nitrate

1.0%

(3)

5.4%

(17)

25.0%

(78)

24.0%

(75)

44.6% (139) Quality logo on the label is

provided

2.2%

(7)

2.6%

(8)

5.4%

(17)

29.8%

(93)

59.9% (187) The nutrient information on the

label is provided

0.3%

(1)

1.6%

(5)

8.7%

(27)

34.0%

(106)

55.4% (173) Visually display actual product

contents (ingredients)

1.0%

(3)

1.0%

(3)

4.5%

(14)

32.1%

100)

`61.5% (192) 100% genuine and safe to

consume

0.6%

(2)

1.0%

(3)

3.2%

(10)

19.6%

(61)

75.6% (236)

(1)

1.3%

(4)

13.1%

(41)

40.4%

(126)

44.9% (140) Durability on the label is provided 0.3%

(1)

1.0%

(3)

4.5%

(14)

22.8%

(71)

71.5% (223)

(3)

5.8%

(18)

19.2%

(60)

44.6%

(139)

29.5% (92) Country of manufacture EBN

products

1.3%

(4)

2.6%

(8)

15.7%

(49)

43.3%

(135)

37.2% (116)

on EBN products and earned highest salary compared to the members

of cluster 1 All the members of this group were non-Malays and they were influenced or encouraged largely by their parents in actual deci-sion making However, subtle differences were found between these two clusters on their employment status, education, age and store they visited to purchase EBN These findings are in line with the result of

a study done by Neeley (2005) and Muralidharan, Rejón-Guardia, and Xue (2016), which showed that friends and families have an influence in individual purchase decision Because of reliance on parents and friends are crucially important in buying EBN products, it is pertinent that the conversational word of mouth among them is far stronger and effective than other types of marketing (Kim, Han, & Lee, 2001; Md Husin, Ismail, & Ab Rahman, 2016; Richardson & Domingos, 2002) Besides, word of mouth can reduce the risk of decision making (Godes & Mayzlin, 2004) and more reliable than information they received from the product provider (Khalid, Ahmed, & Ahmad, 2013)

Collectively, the findings specified the characteristics of buyers that will be useful to build a target list of potential customers and develop new offerings to meet the new needs of customers for EBN products in Malaysia In sum, a market segmentation strategy can help deliver contents that suitable to its particular customers (Bruwer, Roedoger, & Herbst, 2017; Hunt & Arnett, 2004; Kolhede & Gomez-Arias, 2016) Through segmentation, marketers can use this information to tailor their strategy and targeting this specific group of customers that have similar interests and needs Creating a narrow segmentation helps reaching target more effectively with strong and clear message to gain devotion of customers Through this approach, it will provide signifi-cant benefits to marketers and a profitable sales volume to business Suffice to say, market segmentation potentially helps business to grow without significantly increasing marketing expenses

In addition, members from both clusters placed relatively high con-cerns on their health level which reflect the buyers acknowledging the benefits of consuming EBN as additional supplements to their body Although, they were also simultaneously concerned on safety features of EBN as they were aware and feared that their health will deteriorate after consuming the highly purported nutritional product In this sense, EBN producers should establish quality assurance scheme to eliminate the risk with regard to food safety From this perspective, customers’ confidence can be elevated if from harvest to consumption, the EBNs are not contaminated Relatively, it is also suggested that the EBN products should comply with relevant product safety requirements before placing the EBN products in the market