Purchase Intention Model of Malaysian Home-Grown Beauty Product

This study aims to identify factors that contribute to purchase intention of home-grown makeup products. Psychological factors (attitude, perceived behavioural control and subjective norms) and past experience that are based on Planned Behaviour theory are the predictors of home-grown makeup product purchase intention. Cross sectional data collection through survey questionnaire gathered 400 responses from makeup consumers in northern Malaysia. In order to test the relationships, data analyses of the measurement and structural model were performed using Partial Least Squares (PLS-SEM). The structural model test shows a significant positive relationship between past experience (β=0.743, p-value 0.001) and perceived behaviour control (β=0.118, p-value 0.026) towards purchase intention. However, the data shows that attitude and subjective norm has insignificant relationship with purchase intention. This study provides valuable insight into makeup product consumers’ behaviour using samples from northern Malaysia.


Introduction
Home-grown makeup industry Makeup refers to beauty product that is mostly used on face. Face makeup have come into the category of base makeup and point makeup. The examples of makeup product are face powder, foundation, lipsticks, blushers, eye shadow, eye liners and mascara (Matsuoka et al., 2006). According to the Global Islamic Report (2014), in overall beauty products, makeup products are used up to 20% around the world. Euromonitor (2018) reported, in 2017 beauty and personal care industry in Malaysia reported a GDP growth of 5%. This is indicating that Malaysians were more confident in making purchases of various beauty and personal care categories, such as colour cosmetics, skin care and fragrances.
Based on research, country-of-origin (COO) plays a huge influence on consumers' decision to purchase makeup product. More affluent consumers tend to purchase imported products such as L'Oreal, Estee Lauder, NARS, Maybelline, Urban Decay and Bobbi Brown (Baharudin, 2018) because of the established brand name and are globally well known. However, there is some segment of consumers who prefers local products. This is associated with the high spirit to support local or home-grown makeup industry. However, recently the younger consumers are claimed to be influenced by music and film industry in their decision to purchase makeup product. For example, the Korean music and drama series has influenced many consumers around the globe and affected various industries (i.e., fast-food, fashion, accessories). Since the K-Pop trend invaded the film and music industry, younger consumer group had shifted their preference to Korean products. This is due significant influence of the Korean music and film celebrities. Following the trend, therefore, many Malaysian-brand makeup were founded by Malaysian celebrities i.e., Hunnymadu (Madu Cosmetics), Syed Faizal Syed Noh (Syedskillereyes), Nora Danish (OhMost Wanted), Sharifah Nabilla Al-Yahya Syed Sheh (SugarBelle), Vivy Yusof (Duck Cosmetic), Siti Nurhaliza Tarudin (SimplySiti) and Aliff Syukri (D'herbs).
These home-grown makeup entrepreneurs in Malaysia, although operating on a small-scale, some managed to penetrate neighbouring countries markets i.e., Brunei, Indonesia and Singapore. Some managed to be placed at the shelf of established retail stores, pharmacies and global departmental stores in the country (Baharudin, 2018). According to Asia Personal Care and Cosmetic Market Guide 2016, Malaysia's total trade volume for beauty product was about RM8.9 billion in 2015. Based on survey, there are at least 210 manufacturers in Malaysia, and these manufacturers had received the certification of Good Manufacturing Particles and fulfilled the requirement of the ASEAN Guidelines for Cosmetics. Due to huge potentials of the makeup industry, the factor that influenced consumers' purchase intention towards buying home-grown makeup product deserved to be further explored. To the best of the researchers' knowledge there is no study on purchase intention of home-grown makeup product in Malaysia.

Purchase Intention
Based on literature, purchasing intention are influenced by consumers' psychological and beliefs. A belief is the key role that influences consumer to choose and evaluate a product (Keller, 2001). Purchase intention is used in predicting purchasing behaviour of a consumer (Ghosh, 1990). In the context of purchasing, consumers are claimed to be driven by physiological motivation that stimulates their respond which bring them to buy certain type of beauty product (Kim & Jin, 2001). However, decision to purchase differs from one type of product to another, and to make a decision on buying makeup and beauty products is a complex and complicated because it might affect ones' physical appearance (Kim & Seock, 2009). Some other study shows that purchase intention can be examined through perception -which is formed by functional, experiential, and symbolic values (Hung et al., 2011). Nevertheless, past researches show mixed of factors influencing ones' purchase intention.

Significance of Research
The rapid growth in numbers of companies and sales of home-grown makeup products has intrigued researchers' interest; however, there is limited study on purchase intention toward home-grown makeup products. In addition, most of the existing researches that has been done on beauty products industry focuses on organic and natural beauty products or luxury beauty products rather than home-grown makeup products. Limited studies focusing on home-grown makeup, especially in Malaysian context has been done. Furthermore, with limited studies, the factors that influence purchase intention among Malaysian on homegrown makeup products are inconclusive.
Therefore, this study aims to examine the relationship between psychological factors (attitude, perceived behavioural control and subjective norms) and past experience to purchase intention in the context of home-grown makeup product in northern Malaysia. The theory of planned behaviour by Ajzen (1991); Conner (1998) are the basis for the hypothesized relationships. The research finding will broaden our understanding on how does psychological and past experience influenced the purchase intention behaviour for homegrown makeup products consumers, and adds to literature in this area.

Literature Review
Hypotheses development and past researches Predicting purchase intention using Theory of Planned Behaviour (TPB) is widely examined (Kim & Chung, 2011;Kaur et al., 2014;Askadilla & Krisjanti, 2017). Ajzen's (1991) Theory of Planned Behaviour uses attitudes, subjective norms and perceived behavioural control to predict intention with relatively high accuracy. Studies on past experiences as a factor to predict intention to purchase also show positive significant relationship (Weisberg, 2019;Moreira et al., 2017;Barnes et al., 2016;Zhang et al., 2016;Pappas et al., 2014). Kim & Chung (2011) studied on consumer purchase intention for organic personal care products uses TPB constructs (attitude, perceived behavioural control and subjective norms), consumers values (health consciousness, environmental consciousness and appearance consciousness) and past experiences on consumer purchase intention in the case of organic personal care products. Kim & Chung (2011) also studied the moderating effect of perceived behavioural control on the attitude and purchase intention relationship. Kim & Chung (2011) analysed the data using multiple regression analysis from the data of 207 respondents in USA. The findings of Kim & Chung study shows that environmental consciousness and appearance consciousness positively influence attitude toward buying organic personal care products. All the measures of TPB show significant positive relationship in influencing purchase intention. Furthermore, past experiences as a variable to predict purchase intention also show positive relationship. Behavioural control as moderator in the relationship between attitude and purchase intention is significant in Kim & Chung (2011) model of research. Teng & Jamaliah (2013) studied on factors affecting consumer purchase intention toward halal labelled cosmetics and personal care products in Malaysia. This study applied the theory of Reasoned Action (TRA). Six factors that will influence the consumer behavioural intention were determined i.e., past and current experiences of using the products; religiosity; Halal logo and certification; subjective norm; attitude and knowledge. This study was conducted in Malaysia and self-administration survey questionnaires surveys were collected from 573 respondents. Teng & Jamaliah (2013) analysed the data using binary logistic regression to test the relationships between the variables. The findings show that consumer knowledge, past and current experience, subjective norm and religiosity are the factors that contribute to influence the consumers' intention to purchase Halal labelled cosmetic and personal care products. Kaur et al (2014) conducted research in Malaysia in predicting working women purchasing behaviour of Malaysian halal cosmetic products by using theory of planned behaviour. The research used self-administered survey questionnaire to gather the data from 425 working women in the Klang Valley. The independent variables consist of the three main constructs of theory of planned behaviour: attitude, perceived behavioural control and subjective norm to determine the purchase behaviour. Multiple regression technique was used and the finding shows that all of the predictors (attitude, perceived behavioural control and subjective norm) have a positive relationship toward purchase behaviour. Kaur et al. (2014) found that subjective norm is the most important predictors toward the purchase behaviour of working women in Malaysia. Aziz et al (2017) studied on the effects of Theory of Reasoned Action (TRA) on consumers' purchase intention using Partial Least Squares (PLS) approach. This study identified the factors that influence the intention to purchase halal cosmetic products among consumers. The factors are attitude, subjective norm, safety and knowledge. In this research, the data was collected in Malaysia and findings indicated that only three factors have influenced the intention to purchase halal cosmetic products among consumers in Malaysia which are the attitude, subjective norm and knowledge, while safety show no significant relationship towards consumer purchase intention of halal cosmetic products. Askadilla and Krisjanti (2017) studied on green purchase behaviour toward green cosmetic products using theoretical model developed from The Theory Planned Behaviour. The research aimed to examine beliefs (i.e., behavioural belief, normative belief and control belief) influenced toward the three predictors of TPB (attitude, subjective norms and perceived behavioural control) and predict the relationship toward green purchase intention.
The study was carried out in Indonesia and data was collected from Indonesian green cosmetic products consumers' using web-based survey. The 275 responses were analysed using Statistical Package for the Social Sciences (SPSS). The result shows that beliefs would positively affect the three predictors on TPB (attitude, subjective norms and perceived behavioural control). The results also show that attitude, subjective norms and perceived behavioural control has a significant positive effect on green purchase intention. The results also indicate that both green purchase intention and perceived behavioural control has a significant relationship toward green behaviour. Additionally, mediating effect of green purchase intention was shown significant between perceived behavioural control toward green behaviour.
Past study by Yoo and Lee (2011) has been made on asymmetrical effects of past experiences with genuine fashion luxury brands and their counterfeits on purchase intention. This study evaluates the purchase intention toward counterfeit luxury brands (CLBs) and genuine luxury brands (GLBs) by examining the effect of past experiences. The survey was conducted toward 500 respondents of female college students in South Korea. Based on survey data recorded, Partial Least Square (PLS) analysis were used and the finding shows that past experiences with genuine luxury brands (GLBs) are negatively related to purchase intention of counterfeit luxury brands (CLBs), whereas past experiences with counterfeit luxury brands (CLBs) do not relate to purchase intention of genuine luxury brands (GLBs).

Theory of Planned Behaviour Factors
The most common theory used in explaining purchase intention is theory of planned behaviour (TPB) due to the robustness of the theory and applicability to predict purchase behaviour in many contexts such as purchase of organic product (Kim & Chung, 2011;Kaur et al., 2014;Askadilla & Krisjanti, 2017). There were studies on halal cosmetic and personal care products purchase intention in Malaysia and those studies employed theory of reasoned action (TRA) (i.e., Teng & Jamaliah, 2013;Aziz et al., 2017). Other than the factor that is derived from TRA and TPB, Yoo & Lee (2011) and Pappas et al. (2014) tested 'past experience' as predictor of purchase intention. But the context of home-grown makeup product has not being investigated in Malaysia. The theory proposed three predictors for purchase intention: attitude towards the behaviour, perceived behavioural control and subjective norm.
Attitude toward behaviour is defined as an individual's positive or negative feelings (evaluative affect) about performing the target behaviour (Ajzen and Fishbein, 1975). In other word, attitude is the overall evaluations of the behaviour as favourable or unfavourable. An individual is more likely to undertake certain behaviour if that person has a positive attitude toward undertaking the behaviour (Ajzen, 1991). Numerous studies support the positive relationship between consumers' attitudes and behavioural intentions for beauty products. The studies on consumers attitude toward halal cosmetic products show that the results indicate that there is a positive relationship between attitude and intention to choose Halal cosmetic products (Abd Rahman et al., 2015;Brijana & Mursito, 2017). Positive relationship between attitude and purchase intention also has been proven in another study on consumer attitude and purchase intention toward Islamic banks (Nizar, 2013). Juho (2015) studied on attitude toward virtual good purchase versus enjoyment and the finding show Attitude toward buying virtual goods is positively associated with purchase intentions for virtual goods. A study by Long (2014) on consumers attitudes toward blogger's sponsored recommendations and purchase intention based the effect of sponsorship type, product type, and brand awareness show the findings that those consumers have highly positive attitudes toward sponsored recommendation posts that influencing purchase intention based on when products recommended in blog posts are search goods or have high brand awareness. Mohamed (2015) has studied the consumers' attitude toward organic food and the results revealed that the attitude has great importance in influencing the purchase intention to consumers.
According to Ajzen (1991) perceived behavioural control is the perceived ease or difficulty of performing the behaviour. Individual's perception of the extent to which performance the behaviour is easy or difficult (Ajzen, 1991;Conner et al., 2001). Thus, those who perceived a higher degree of personal capability and control tend to have stronger behaviour intention to engage in certain behaviour (Ajzen, 1991). Perceived behavioural control refers to the degree of control that a person perceives over performing the behaviour (Chen, 2007;Kang et al., 2006). Thus, people who perceive a higher degree of personal control tend to have stronger behavioural intention to engage in a certain behaviour (Ajzen, 1991). In particular, when people believe they have more resources such as time, money, and skills their perceptions of control are high and hence their behavioural intentions increase. Suzanne (2014) has studied to examine the determinants of intention to purchase travel online and the finding showed that perceived behavioural control positively influences intention to purchase travel online. A study on young consumer's intention toward purchasing green products show that perceived behavioural control is significantly influencing the consumer's purchase intention to buy green products (Rambalak, 2015). Another study by Chen (2013) on predicting consumers intention to visit green hotels using Theory of Planned Behaviour proved that perceived behavioural control is one of the positive significant factors in influencing the consumers intention. However, a study by Edmund (2016) on non-compliance in national parks using extension of the theory of planned behaviour model with pro-environmental values showed that the role of perceived behavioural control is not significant in this study.
Subjective norm refers to a person's perception that most people who are important to him think that he should not perform the behaviour in question (Ajzen and Fishbein, 1975). Subjective norms usually evaluate the perceived social pressures to perform or not perform a particular behaviour (Ajzen, 1991). Subjective norm reflects one's perceived social pressure to perform a certain behaviour. If consumers believe that significant others think home-grown beauty products are good, then consumers will have more intention to buy these products. On the contrary, if consumers believe that people important to them have negative opinions toward the products, then consumers will have lower intention to purchase home-grown beauty products. Following the Theory of Planned behaviour, a study by Chi (2014) studied on the decisional factors driving toward organic food consumption show that subjective norm played significant role as one of the factors in organic food consumption. A study on predicting Iranian students' intention to purchase organic food show the findings of the study that subjective norm was not a significant predictor toward purchase intention in this study (Masoud, 2014). Lim (2016) studied on factors influencing online shopping behaviour in Malaysia and the results show that subjective norm is insignificant factor in influencing online shopping behaviour. A studied by Rambalak (2014) investigated consumer's intention to purchase organic food in India using TPB also show that subjective norm failed to show significant influence toward purchase intention of organic food. Justin (2015) studied on predicting green product consumption using theory of planned behaviour and reasoned action and it show that subjective norms does not significantly predict purchase intention. Contradict with TPB, a few studies show insignificant results toward intention. Based on these reviews this research employs TPB factors to predict purchase intention behaviour.

Hypothesized Relationships
Theory of planned behaviour (TPB) has been adopted to study the factors that contribute to the purchase intention toward home-grown beauty product in Malaysia. Additional factor added as the factor is past experience. Therefore, to examine the four factors (attitude, perceived behavioural control, subjective norms and past experiences) toward purchase intention of home-grown makeup products the following hypotheses are proposed: H1. Consumers' attitude has a significant positive influence on their intentions to buy homegrown makeup products. H2. Consumers' perceived behavioural control has a significant positive influence on their intentions to buy home-grown makeup products. H3. Consumers' subjective norm has a significant positive influence on their intentions to buy home-grown makeup products. H4. Consumers' past experiences with home-grown makeup product has a significant positive influence on their intentions to buy home-grown makeup products.
The suggested framework based on Theory of planned behaviour by Ajzen with extensional of additional factor of past experiences (Conner, 1998) is shown as below:

Participants
The participants in this research are makeup product consumers in northern peninsular Malaysia. In order to approach the consumers, emails were sent to the home-grown makeup companies requesting for their customers list. Based the list provided by the companies, researcher request for the company to help in approaching their customers and distribute the survey questionnaire. Most of the home-grown companies using social network sites to get to their customers and the survey were sent through a link at the companies' social network sites e.g., Facebook and Instagram. Furthermore, the home-grown makeup product companies employ drop-ship agents to sell their products, therefore emails and WhatsApp messages was sent to the drop-ship agents to help in distributing the survey link. Link to the survey questionnaire were created for this research.
The targeted participants for this research are makeup product consumers at the northern Malaysia. Therefore, fours states consist of Penang, Kedah, Perak and Perlis were the geographical boundaries of this research. The rationales for limiting the survey to these four states only are: 1) to control the factor of points of access to home-grown makeup product. Usually, home-grown makeup products are distributed based on regions; and 2) to minimize the socio-economic divide among the consumers.

Sampling
In this study, the minimum sample size is determine using G*Power calculator. It is a tool that allows high-precision power and sample size analyses (Buchner et al., 1997) because it computes power values for given sample sizes, effect sizes, and alpha levels (post hoc power analyses), sample sizes for given effect sizes, alpha levels, and power values (a priori power analyses), and alpha and beta values for given sample sizes, effect sizes, and beta-alpha ratios (compromise power analyses). G*Power was chosen because of its ease of use and the wide range of study designs for which it caters. Using G*Power, sample size and power can be estimated or checked by those with relatively little technical knowledge of statistics.
The sample size is 129 derived from G*Power application for calculation for accurate results to achieve the objectives and hypothesis. Data collection process begin with by doing research through online to find list of makeup products companies available in Malaysia.

Measure of Constructs
Constructs of research are measured using survey questionnaire. The survey questionnaire included questions/items on attitude toward buying home-grown makeup products, subjective norm, perceived behavioural control and past experiences to home-grown makeup products purchase intention. There are two sections in the questionnaire: the first section contains questions about demographic and socioeconomic characteristics of the respondents. The second section contains questions on: attitude toward home-grown makeup product, subjective norm, perceived behavioural control purchase intention and past experiences. All questions were measured on a five-point Likert scale. The questionnaires are adapted from previous research and summarized in Table 1.

Table1: Questionnaire Design Item
Adapted from Attitude (ATT) For me, buying home-grown make up products is pleasant. Conner et al. (1999) For me, buying home-grown make up products is good. Conner et al. (1999) For me, buying home-grown make up products is beneficial. Conner et al. (1999) For me, buying home-grown make up products positive.

Subjective norm (SN)
People who influence my decisions would approve of me buying home-grown make up products. Bansal & Taylor (2002) People who are important in my life would approve of me buying home-grown make up products. Bansal & Taylor (2002) People who influence my behaviour would encourage me to use home-grown make up products. Lin (2007) People who are important in my life would encourage me to use home-grown make up products.

Perceived behavioural control (PBC)
For me buying home-grown make up products would be possible. Ajzen (2002) If I wanted to, I could easily buy home-grown make up products. Ajzen (2002) I believe that I can buy home-grown make up products.

Purchase Intention (PI)
If home-grown make up products are available, I would buy it. (2008) It is likely that I will purchase home-grown make up products. (2008) I plan to buy home-makeup make up products. Lin (2007)

Data Analysis
To test the convergent validity and discriminant validity reliability test, firstly the Indicator Reliability is measured using Factor loading. Factor loading suggested value is higher than 0.6. Factor Loadings represent the absolute contribution of the indicator to the definition of its latent variable. (Fang et al., 2014;Han et al., 2015b;Setia et al., 2013;Urbach & Ahlemann, 2010;Wang et al., 2013). Composite reliability (CR) is used to test convergent validity as composite reliability (CR) may lead to higher estimate of true reliability. According to Urbach and Ahlemann (2010), the desirable value for CR is higher than 0.7 to use for assessing internal consistency. A commonly applied criterion of convergent validity is the average variance extracted (AVE) proposed by Fornell and Larcker (1981). An AVE value of at least 0.4 indicates the model are able to explain more than half of the variance of its indicators. This demonstrates sufficient convergent validity (Garson, 2016;Urbach & Ahlemann, 2010). It measures the percent of variance captured by a construct by showing the ratio of the sum of the variance captured by the construct and measurement variance (Gefen et al., 2000). Next, discriminant validity involves the degree to which the measures of different constructs differ from one another. In PLS, two measures of discriminant validity are commonly used: Cross loading criterion and heterotrait-monotrait (HTMT). This is because HTMT are able to detect the lack of discriminant validity better than Fornell-Larcker proven from a simulation study (Jörg Henseler et al., 2014). The suggested HTMT ratio is 0.85.
To measure the structural model evaluation using partial least square (PLS) the assessment of the model's quality is based on its ability to predict the endogenous constructs. The following criteria facilitate this evaluation: Coefficient of determination (R 2 ) (Urbach & Ahlemann, 2010), predictive relevance (Q 2 ) (Evermann & Tate, 2014), and path coefficients (Garson, 2016). According to Evermann & Tate (2012), it is recommended to use redundancybased blindfolding to assess the predictive relevance of one's theoretical/structural model. The suggested value of Q 2 > 0.5 indicates a predictive model. R 2 is the measure of the proportion of the variance of the dependent variable about its mean that is explained by the independent variables (Gefen et al., 2000).

Demographic Profiling
A total of 180 usable data was collected from the survey conducted through online. The demographic profile of the respondents has been identified in this study in Table 1. There are seven questions were asked in the demographic section in the questionnaires. The demographic questions consist of gender, age, race, state, level of education, occupation and income status. Number of past purchases of home-grown makeup products from the last three months by the respondents also were recorded for the research. From the findings, the total respondents of this study are 100 per cent Malaysian and female. The demographic profile of the respondents has been summarized below in Table 2. 16.7 7-9 6.1 10-12 3.3 More than 15 1.7

Measurement Model Evaluation
Two types of validity have been measured to evaluate the measurement model in this study, the convergent validity and discriminant validity. The convergent validity was examined the factors loadings, average variance extracted (AVE) and composite reliability (CR). The factors loading for all items are higher than 0.4 as suggested by. To measure the reliability, the composite reliability was measured. The composite Reliability show for attitude (0.962), Perceived Behavioural Control (0.896), Past Experiences (0.943), Purchase Intention (0.943) and Subjective Norm (0.922) which are higher than 0.7. The Average Variance Extracted (AVE) values for attitude (0.865), Perceived Behavioural Control (0.742), Past Experience (0.847), Purchase Intention (0.878) and Subjective Norm (0.749) which are higher than 0.5, the minimum value suggested. The summary of the assessment of the measurement model is recorded in Table 3. The Heterotrait-Monotrait (HTMT) was measured to check on the discriminant validity. The measure was tested according to criterion of comparing between the constructs and the square root of AVE for the construct. As suggested by Garson (2016) all HTMT value are less than 0.85, proven that the discriminant validity is established since all discriminant validity are satisfied. Table 3 show all the correlation between the constructs. Overall, the measurement model demonstrated adequate convergent validity and discriminant validity satisfied the construct validity for the measurement model of PLS. Thus, from the findings all four independent variables (attitude, perceived behavioural control, subjective norms and past experiences) had significant contribution in influencing purchase intention toward homegrown makeup products.

Structural Model Evaluation
The structural model is measured by calculating R 2 , path coefficient, predictive relevance Q 2 and p-value of the constructs. The structural model was analysed using PLS Bootstrapping techniques by using 2000 samples. The R 2 result is 0.815 showing there is 81.5 per cent of variance in Purchase Intention. Predictive relevance Q 2 was analysed using blindfolding technique to analyse as an additional model fit assessment. The Q 2 of the study show 0.674 which is higher than 0.5, the suggested value showing by Evermann & Tate (2012) that the model has sufficient predictive.
From the finding in this study, only two factors show positively significant results toward purchase intention. Perceived behavioural control (β= 0.118, P-Value: 0.026) and Past experience show (β= 0.743, P-Value: 0.000) was positively related to purchase intention toward home-grown makeup products. This supported hypothesis H2 and H4. While, attitude (β= 0.106, P-Value: 0.153) and subjective norm show (β=-0.028, P-Value: 0.260) show insignificant relationship toward purchase intention to home-grown makeup products. Thus, hypothesis H1 and H3 is not supported. The summary of the result of the research hypothesis H1, H2, H3, and H4 are recorded in Table 5. The structural model consists of factor loadings of each indicator, R 2 , Predictive relevance Q 2 , path coefficient, p-value, and indicator of hypothesis supported is summarized in the figure 2 below.

Discussion
The purpose of this research is to study the predictors of purchase intention i.e., attitude, perceived behavioural control, subjective norms and past experiences. The findings show that there is a significance positive influence between perceived behavioural control and past experiences toward purchase intention. The findings were consistent with past research regarding purchase intention (i.e., Kim & Chung, 2011;Kamaljeet et al., 2016;Askadilla and Krisjanti, 2017;Antonio, 2015;Stuart, 2015;Zhang, 2016;Ilias, 2012;Phuah & Wan, 2016;Aziz et al., 2017;Yoo & Lee, 2011). However, attitude and subjective norm was found to be insignificant.

Insignificant relationships
Attitude refers to ones' overall evaluations of a behaviour as favourable or unfavourable. In this research, consumers' attitude toward home-grown makeup product are measured by items: 'finding the products are pleasant', 'good', 'beneficial' and 'positive', with high loadings for each item's 0.921, 0.937, 0.930 and 0.932 respectively. However, based on the finding it shows that attitude as a construct has insignificant relationship to purchase intention toward home-grown makeup products. This finding is not consistent with past researches i.e., Azmawani, 2015;Vita, 2017;Nizar, 2013;Juho, 2015;Long, 2014;Mohamed, 2015. Subjective norm (SN) refers to one's perceived social pressure to perform a certain behaviour.
In this research SN found to be in significantly affected the purchase intention toward homegrown makeup products. The high loadings of each indicator are 'people who influence their decisions approval' (0.765), 'important people in their life approval' (0.871), 'encouragement from people who influence their behaviour' (0.920), and 'encouragement from people who are important in their life' (0.898), these indicates that the measurement items for SN construct is good. However, finding of the structural model shows SN as a construct did not influence consumers' purchase intentions in the context of home-grown makeup products. This finding is consistent with Masoud, 2014;Lim, 2016;Rambalak, 2014;Justin, 2015. However, this finding is contradicted with Chi, 2014 and Theory of Planned Behaviour by Ajzen (1994). The probable explanation is to purchase home-grown makeup product social pressure is not taken into consideration. This is might be due to the loose social pressure in this context of purchasing home-grown makeup, and their dependency on approval from internet and social media related to home-grown makeup products.

Significant Relationships
Based on researches, past experiences (PE) make a good predictor for purchase intention. This research found that past experience has a significant positive influence on purchase intention. It reveals that the construct of PE which is measured by 'feel good and satisfied', 'will recommend it to their friend' and 'no hesitation' with each item's loadings are 0.928, 0.900 and 0.932 are good indicator of past experiences. This result is consistent with previous studies on past experiences such as Yoo and Lee (2011); Jacob (2019); Antonio (2015); Stuart (2015); Zhang (2016); Ilias (2012). Therefore, to predict future purchase intention for homegrown makeup, first experience of the customer is important.
Perceived behavioural control (PBC) refers to ones' perception of the extent to which performing a behaviour is easy or difficult. In this research, PBC is measured on three items: 'possible to purchase', 'easily will purchase' and 'the ability to purchase' with each item show high loading of 0.839, 0.886 and 0.858 respectively. PBC as a construct showed significant positive relationship to purchase intention toward home-grown makeup products. The finding is consistent with past research such as: Suzanne, 2014;Rambalak, 2015;Chen, 2013;Edmund, 2016.

Research Implications
Hence, these significant findings have several important implications for the entrepreneurs and Small Medium Enterprise (SMEs) in Malaysia in formulating strategies to further enhance their makeup products which are as follow: Improving customers' perceived behavioural control The perceived behavioural control showed significant positive relationship to purchase intention toward home-grown makeup products implies that marketing strategy of 4Ps (Products, price, promotion and places) has influenced customers' perceived behavioural control. Good product describe by attractive packaging does promotes perceived behavioural control. The pricing strategy in marketing also require attention in order to stimulate the consumers' perceived behavioural control toward purchase intention of home-grown makeup products. A variety of communication programs such as media advertisement and social media advertisement will positively affect the perceived behavioural control of consumers. Also, promoting by endorsement using celebrity and social media influencers also shows positive effect on the perceived behavioural control. The entrepreneurs of homegrown makeup products should place their products in well-known makeup store such as Sephora, Watsons and Guardian for easy accessibility to the consumers. Easy accessibility to consumers will encouraged the user to have positive perceived behavioural control toward purchase intention on home-grown makeup products Also, following the technology, online makeup selling through website also is encourage. User-friendly interface and simple steps for the consumers to purchase will give positive effect toward the consumers' perceived behavioural control.

Improving Customers' Experiences
The past experience does have positive significant effect on purchase intention of homegrown makeup products. From the findings, if past experience of home-grown makeup products is positive, consumers are found to have no hesitation in purchasing the homegrown makeup products, followed by feel satisfied of past purchased toward home-grown makeup products and will recommend it to their friends. Past experiences usually will lead customers' satisfaction, which means if consumers have good past experience with the products it will lead to the customers feel satisfied with the products. Next, satisfaction with the products will lead to consumers put trust on the products. Once the customers have trusted the products, customer will purchase again the products and will recommend the products to others. In order for the customers has good experience with the products, entrepreneurs and small medium enterprises (SMEs) need to ensure the products has what consumers need. Past experiences not only regarding the products but also the surrounding factors such as the process of purchasing the products and the accessibility of the products. Other than the products details need to be considered such as price, attractive packaging, practicality of the products, the whole experiences of purchasing such as interaction while online purchasing, accessibility of the home-grown makeup products and other offering from the products also should be taken care of by the entrepreneurs. As suggestion, the entrepreneurs should make more research and development in enhancing their makeup products to suits Malaysian skin and Malaysia's weathers. Also, the entrepreneurs and SMEs can create a loyalty program and providing free samples of new products toward the members, this can give good experiences toward the consumers.
For future research, study on purchase intention toward home-grown makeup products can be done as in a whole Malaysia for better generalizing and understanding to bigger group as current study has only been conducted in Northern region of Malaysia. This is to study the consumers purchase intention toward home-grown makeup products in Malaysia as a whole as different region might have different results of the research because different region might have their own market and consumer purchase behaviour. Comparison can be made based on regions and further study can be made based on the results. Secondly, this study is only focusing in makeup products only. More future research could be done in bigger context such as beauty products as a whole. This is because the finding from this research may not be the same to other home-grown beauty products such as skincare. Also, future study can use halal context as a measure of purchase intention toward home-grown beauty products. This can help the entrepreneur of makeup product to understand the influence of halal concept toward purchase intention.

Conclusion
In conclusion, past experience as a factor of purchase intention as proposed by Conner and Armitage (1998) and theory of planned behaviour by Ajzen (1991) formed a good model to explain purchase intention of home-grown makeup product. The structural model explained 81.5% of the purchase intention toward home-grown makeup products and the measurement model also showed adequate validity and reliability, with Composite Reliability (CR) values of all variable are higher than 0.4 and all the (AVE) for each variables show values higher than 0.5. This research can be replicated with some improvement such as to include more respondents to increase the findings' representation.