CHAPTER 2
LITERATURE REVIEW
In the study, articles from several journals are be reviewed, of the articles are related to awareness of consumer in urban agriculture. The review of previous studies can provide researchers with information about models used, variables, techniques for analysis and result of related studies. The review can be used as guide lines for the present study. 2.1 Related Past Studies
Consumer awareness is the understanding and knowledge that a buyer should have of his rights as a customer. The awareness is very important for the buyer since it permits him to get the most from what he buys. To develop or start new practices need lot of effort from many side to ensure it can be done, not only from supply side but demand side play an important role in this. Even there is lot of benefit from urban agriculture, we still need to know if public know about urban agriculture or it related practices.
The result from study of consumer awareness of organic products in Spain, T. Briz and R. W Ward (2009), show that education was a major contributor factor that influencing the level of awareness. From this conditions, estimated awareness levels ranged from a low 29% to high near 71%. The most profound result is that while there is positive link between awareness and actual buying point. A research in India, Sushil Kumar (2011) give the same result which indicated education plays important role to enhance awareness among people. 64% of respondents with awareness on organic foods have higher educational level (degree holder and higher) compared to 22% in case of unaware consumers.
Education background does play important role affecting awareness level of consumer. But in study about consumer awareness, motivations and buying intention of eco-friendly fast moving consumer goods by Dr. H. C. Purohit show the young generation is more likely to aware about this situation may because of the high promotion and campaign with various medium.
In marketing, psychology, and sociology, the perceptions of (familiarity with or knowledge about) an ideal, product, or service play an important role in an individual’s choice (preferences or non-choice) of that particular idea, product, or service. In other words, product or new technology preference is influenced by perception, familiarity, and/or knowledge of the product. The more favorable the perception, the greater the likelihood from among similar alternative.
Since the study about consumer perception towards urban agriculture is still few, I refer to several other related articles such as consumer perception toward organic food. Gemma C. Harper and Aikaterini Makatouni (2002), did the study about consumer perception and the result show that consumer have similar understanding toward organic but they have divergent understanding between organic food and free range production. The study also show their perception is affect by their belief toward health issue and lifestyle. Another study by Samuel Bonti-Ankomah and Emmanuel K Yiridoe (2006) say that consumer perception is mostly depend on knowledge and awareness level about organic product. If the skepticism about organic products stemming, perceptions about the appeal and inherent characteristics of organic may translate into actual demand.
Another study in Bangkok about consumer perception toward organic food by Birgit Roitner-Schobesberger, Ika Darnhofer, Suthichai Somsook, and Christian R. Vogl(2008) show that older, highly educated and high income more preferable to purchase organic product. The study also indicated that lack of knowledge toward specific subject restrain the consumer from purchase or try an organic product.
Beliefs and perceptions are highly subjective concepts (Fishbein and Ajzein, 1975), because they reflect opinions about the objective state of the world. Although in reality such perceptions may or may not be true, the individual who holds the perception thinks that it is true. Given Lancaster’s (1966) conception that consumers demand bundles of product characteristics, perceptions about particular (desirable) characteristics of organic food can influence a buyer’s choice. Studies on consumer perceptions about organic versus conventionally produced foods therefore attempt to determine what consumers think is true.
Attitude is conceptualizes in many ways from being a state of readiness for mental and physical to prefer and evaluate object or aspect in favorable or unfavorable manners. Attitude can be considers as learned and summary evaluation that influences thoughts and actions. (Dawes, 1972). Several collective evidence from environmental psychology and landscape research show attitude towards the environment is multi-dimensional theory with common orientations, demographics, knowledge, and perspective (Blake, 2001; Schultz and Zelezny, 1999).
A study in Sweden by Maria K. Manusson, Anne Arvola, and Ulla-Kaisa Koivisto Hursti (2001) in attitude towards organic food resulted that women demonstrate more positive attitudes to buy organic food. The result also indicate that people from urban area more likely to buy an organic food to increase their health. They also prefer organic food because it is fresher and better taste. Another study by Shivanand Balram and Suzana Dragi’ cevi’ (2004) show that attitude towards urban green spaces is a multi-dimensional concept. Two dimensions being support in quantitative phase of analysis where “behavior” toward urban green spaces and “usefulness” of urban green spaces. The result also show the connection between their home areas and their attitude toward urban green spaces. People in city more likely to aware and know and willing to try green space concept. We can see the pattern here where people who from urban area more willing toward new things. They also more prefer if something is good for their health.
There is another study by Anna Saba and Federico Messina (2002) in Rome, Italy about consumer attitude toward organic food; risk/benefit perception prove that consumer want food that is safer and they know risk associated with pesticide in fruit or vegetables. The result also indicate that ‘old people age > 60’ prefer conventional food as long it is cheap. This contradict opinion from others study show that public attitude can be differ with age constraint. Differences between knowledge associate with age level really affect their attitude towards organic food. Consumer attitudes toward organic food product being influence by their positive attitudes that want more nutritious and healthy food compare conventional. Other that health factor, influence from close friend/family/relative also play big roles in their attitudes toward organic food.
Adoption process is the series of stages which consumers determine to do and practice new technology or become regular purchaser of a product/service. The stages start from hearing about product to finally adopting it. There five main stage in adoption process: awareness, interest, evaluation, trial, and adoption/rejection. Never confuse adoption process with diffusion process. Diffusion process is a group of phenomena which suggest how an innovation spreads among consumers. Adoption process is a subset of diffusion process (Everett Rogers, 1963)
Efstratios Loizou, Anastasios Michailidis and Irene Tzimitra-Kalogianni (2009) make a study about drivers of consumer’s adoption of innovative food make a finding that consumers tend to adopt new innovative food products suggest that can helped them to drive real food safety, health, economic and environmental gains. Furthermore, the four classes were found to differ in terms of gender, marital status, income, education and number of children. The increased level of disposable income along with exposure to innovative food products may well explain such a differential innovative food product adoption. A further finding is that innovative food products’ adoption is significantly related to factors such as ‘annual income’, ‘classification of consumers’, ‘knowledge of organic and own label products’, ‘age’ and ‘education’ with ‘number of children’ being a supporting factor. The findings support Rogers’ (1995) socio-economic generalizations about early adopters.
Research about organic food adoption decision by New Mexico groceries by Pony E. Ireland show most of the store owner did not want to carry organic food in their store because of the shortage of supplies and being follow by higher price compare to conventional food. Even though the demand was pretty high, lack of supply and little store that carries organic food can become one of the major problem for consumer who willing to adopt organic food in their daily life.
Even if consumer willing to make a change but if the supply is limited, it will be hard for them to adopt new product/service. Socio-economic play important role in decide consumer to adopt a system or not. Since adoption process will go through several process, consumer will be more convince if the product/service come with prove and example.

2.2 Methodological Issue
In order to measure awareness, perception and attitude and adoption level of consumer toward Urban Agriculture in this study, several model methodology been revise. There is 6 type of different model that constantly been use in this kind of study; Fishbean and Aizen, The Theory of Reasoned Action (1975, 1980), the theory of planned behavior, Icek Ajzen (1985), Multinomial Logistic Model; conditional maximum entropy model or MaxEnt Model for short, Logit Model by Joseph Berkson in 1944. Poisson Regression Model, Structural Estimation Model (SEM), Kruskal-wallis test, and Probit Model.
A study in India by Sushil Kumar (2011), consumer awareness and attitude toward organic food use Logistic Regression Model to analyze the factors affecting consumer’s awareness. The empirical model estimates the relationship between socio-demographic profiles of the consumers and their food safety concerns in influencing the awareness level. Result indicates education and stream of education of the consumer have significant role in affecting awareness level. Likewise, consumer who more concern about food safety and have access ICT based more likely to have better awareness on organic foods. The empirical model estimates the relationship between socio-demographic profiles of the consumers and their food safety concerns in influencing the awareness level.
Other study in Swedish, Attitudes towards organic foods among Swedish consumers by Maria K. Magnusson, Anne Arvola and Ulla-Kaisa Koivisto Hursti (2001), they use Theory of planned behavior (TPB) (Ajzen,1988,1991) to design the questionnaire. The questionnaire consisted of two section; first section two know the attention or perception to buy consumer toward organic food and second section is to study consumer attitude to buy the products. The findings show the majority demonstrated positive attitudes towards buying the four organic target foods. Between 46 per cent and 67 per cent of all respondents agreed that it is quite or very good, wise and important to buy organic and most of them is from consumer who aware about organic food and know the benefit of it. The TPB model seeks to explain behaviors, goals and outcomes that are not entirely under the control of the person (Ajzen, 1991). It assumes three conceptually independent determinants of the intention to perform a certain behavior (Ajzen, 1991). There is three determinant that being measure which the attitude is toward behavior, subjective norm and concerns perceived social pressure to perform or not, and lastly is the proportion of perceived behavioral, and it is assume to reflect pas experiences (Ajzen, 1991). TPB usually use in present study as basic to design the questionnaire with aim to present descriptive data and differences between groups differing in demographic variables. In this study, they also use statistical method to analyze descriptive data. Fisher’s PLSD-test was used for post hoc comparisons. Two-tailed t-test (unpaired) and one-way analysis of variance was used for statistical analysis. The Fisher’s PLSD test takes the square root of the residual mean square from the ANOVA and consider that to be the pooled standard deviation. We compared the sample size of two groups by computes a standard error of difference between those two means. Using this test, the result was women, university educated people and young respondent had more positive attitudes than their comparison groups.
A study by A. Saba, F. Messina on attitude towards organic food (2003) use Kruskal-Wallis one-way analysis of variance to study the Differences between groups of subjects on the mean ratings of the components trust, perception of risks and benefits. Differences between groups of subjects on the frequencies of gender, classes of age, education and geographical area were tested using the chi-square test (Fienberg, 1977). Cronbach’s alpha coefficients of internal reliability were assessed for attitudes, and beliefs responses multiplied by the corresponding evaluations (Cronbach, 1951). Cronbach’s alpha usually use to measure internal consistency and show how closely related a set of item as a group. This test is not a statistical test but more to coefficient of reliability or consistency. Cluster analysis (Chatfield & Collins, 1983) based on the average ratings of the six beliefs was applied in order to identify profiles of beliefs towards the consumption of organic fruits and vegetables. The hierarchical method was used specifying the maximum number of clusters allowed. Another study by Samuel Bonti-Ankomah and Emmanuel K Yiridoe (2006) also use Kruskal-Wallis and Cronbach’s alpha coefficients to analyze the data and result. The hierarchical method of was used specifying the maximum number of clusters allowed.
H Vermeulen, JF Kirsten, TO Doyer, and HC Schonfeldt make a study about attitudes and acceptance of South African urban consumer toward GMF (2005) use conjoint of cluster analysis and structural equation model (SEM) in their research. Conjoint of two type of analysis give more accurate result and make the analysis more specific in this case research objectives in conjoint model is evolved around relative importance of a specific attribute and whether clusters of respondents could be found with similar pattern of importance. SEM is very general and largely confirmatory than explanatory make researcher more likely to use SEM to determine a validity of a model. Structural equation models (SEM) estimated the causal relationships between the variables. SEM was used as confirmatory analysis, which means the model is based on a priori information about the data structure in the form of a specified theory or hypothesis or knowledge. Structural Equation Models (Bollen, 1989; Hayduk, 1987; Jo¨ reskog & So¨ rbom, 1998) are also known as analysis of covariance structures due to the implications of simultaneous regressions that are studied primarily at the level of correlations or covariance’s.
Multinomial and logit model being used in another study about consumer awareness, attitude and perception toward organic food. A study by T. Briz, R.W. Ward (2009) show the result of probability for each respondent is above average which mean any variable or combination of variables were controlled and the average probabilities being derived based on controlled variables. The responses towards organic food are mutually exclusive and exhaustive, once the probabilities for two awareness levels are estimated the third level is predetermined. Logit regression is a linear-regression generalization for the case when the independent variable is nominal. Main advantages of logit regression is it simplicity and a relatively high speed of model generation, but unsuitability for solving essentially nonlinear problems.
Actual use of organic foods is a binary response hence, a probit model provides a useful way for estimating the likelihood of actually buying organically grown products. While awareness is based on consciousness, consumption requires an explicit buying commitment that should be influenced by price and appropriate measure of the organic quality. Since probit model is one of the best model to measure the likeness of the variables. There are several type of model estimation under probit model. The one that being use in the study is Gibbs sampling method. Gibbs is possible because regression models typically use normal prior distributions over the weights, and this distribution is conjugate with the normal distribution of the errors. The final result show the likelihood of purchasing organics should decline with the expectation of higher prices and should increase with a similar expectation of greater nutritional value. Other than Gibbs sampling method, there’s also another technique which is Berkson’s minimum chi-square to estimate unobserved quantities based on observed data. In this kind of analysis, one rejects a null hypothesis about population distribution if a specified test statistic is too large. Test statistic of large sample size usually have roughly a chi square distribution. If chi square estimation is minimum, the value of parameters for test also small as the estimation.
There many models out there that we can use to analyze and measure the result from SPSS in order to get the exact result and has a better understanding about it. But there is some major argument between researcher weather to use logic model or probit model. Since this two model does completely different thing. Pearson (1900) showed that that if multivariate normal data were generated and thresholded to be categorical, the correlations between the underlying variables were still statistically identified - these correlations are termed polychoric correlations and, specific to the binary case, they are termedtetrachoric correlations. This means that, in a probit model, the intraclass correlation coefficient of the underlying normally distributed variables. Multinomial logit models usually more suit and faster to use on larger problem, multiple analysis or larger data set. Logistic regression definitely has the advantage in interpretability of the parameters.