- Email: [email protected]
Exploring Consumers’ Attitudes towards Wood Products that could be Derived from Transgenic Plantations in Greece
Available online at www.sciencedirect.com
ScienceDirect Procedia Technology 8 (2013) 554 – 560
6th International Conference on Information and Communication Technologies in Agriculture, Food and Environment (HAICTA 2013)
Exploring Consumers’ Attitudes Towards Wood Products that could be Derived from Transgenic Plantations in Greece Lambros Tsourgiannisa,b*, Vasiliki Kazanab, Anastasios Karasavvoglouc, Michael Nikolaidisc, Giannoula Florouc, Persefoni Polychronidouc b
a Region of East Macedonia and Thrace, Regional District of Xanthi, 67100 Xanthi, Greece Department of Forestry and Natural Environment Management at Drama, Kavala Institute of Technology , Drama, Greece c Department of Accountancy, School of Business and Economy, Kavala Institute of Technology, 65404, Kavala, Greece
Abstract This paper aims at exploring consumers’ purchasing behaviour towards wood products that could be derived from transgenic plantations. Field interviews were conducted and analysis was performed on a sample of 231 consumers, who declared that they would buy such wood products. Principal Components Analysis indicated that the main factors affecting consumers’ purchasing behaviour towards those products are promotion issues, product features and labelling. Through cluster techniques 3 groups of consumers with similar consumption behaviour were identified: ((a) Consumers influenced by marketing issues, (b) Opportunists (c) Consumers interested in product features. Discriminant Analysis was performed to predict cluster membership and a non parametric test to profile each consumer group according to personal characteristics and attitude towards establishment of transgenic forest plantations. © 2013 byby Elsevier Ltd.B.V. © 2014 The TheAuthors. Authors.Published Published Elsevier Selection and under responsibility of The Hellenic Association for Information and Communication Technologies in Agriculture Selection andpeer-review peer-review under responsibility of HAICTA. Food and Environment (HAICTA)
Keywords: Consumer behaviour; wood products; Forest Transgenic Trees.
* Corresponding author. Tel.: +30 6972084118. E-mail address: [email protected]
2212-0173 © 2013 The Authors. Published by Elsevier Ltd. Selection and peer-review under responsibility of The Hellenic Association for Information and Communication Technologies in Agriculture Food and Environment (HAICTA) doi:10.1016/j.protcy.2013.11.078
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560
1. Introduction Transgenic forest trees are biotechnology products, that is, they are organisms Genetically Modified (GM) or Genetically Engineered (GE) through the insertion of DNA from other species in order to manifest specific traits (www.forestguild.org). Such traits mainly concern higher growth rates, lignin reduction and increased resistance to herbicides or forest pests. At the current state of art there are five main domains of potential application of GM trees: basic and applied research, commercial production, bio-remediation, ecological restoration and response to epidemics. Much discussion has been geared by academics and the industry on the potential of GM forest trees to help meet projected demand increases for wood products worldwide, through increased growth and productivity and decreased cost of wood fiber production [5, 18, 3, 16, 2, 9, 6, 11]. However, with the exception of China, no market for GM forest trees currently exists in Europe, the USA or other parts of the world, despite that over the past 20 years through confined transgenic forest tree field trials and millions of hectares of commercial transgenic crop fields much scientific experience has been gained [8]. The main reasons for the absence of such markets are the very strict regulatory frameworks regarding license granting for establishment of GM plantations, particularly due to a number of potential public concerns. These include the potential for spread of antibiotic or herbicide resistance genes to other non-target species from GM trees; the potential for long – distance pollen spread over many years from long – lived trees, the potential for adverse effects on biodiversity from transgenic forest plantations and any unexpected effects [7, 15, 6]. Also, GM forest plantations might generate profit for certain actors in the private sector, while poorer communities will be further marginalized [17]. The regulation framework of transgenic forest trees in use differs between countries. In the European Union regulations aim at providing high levels of protection for human health and the environment, while ensuring consumer interests and the promotion of competitive market. Socio-economic information of transgenic forest plantations in Europe is completely lacking. Evidence from other countries, such as the USA, indicated that the industrial consumers of wood products, who use wood as an input to production, are generally enthusiastic about transgenic forest trees, contrary to the attitude of consumers towards final wood products, who appear more concerned [14]. Although wood products from transgenic plantations are unlikely to be in the market for another 10-15 years, scientifically based information on the anticipated attitudes of consumers is extremely important both for developers as without the expectation of viable market the developments and investments are unlikely to be forthcoming and for policy makers, who need to respond adequately through regulation tools and programs. Hence, this paper presents results of a study, the first of its kind in Europe to the authors’ knowledge, initiated in Greece to explore the attitudes of consumers towards wood products that could be derived from transgenic forest plantations and profile them according to personal characteristics and their attitude towards establishment of such plantations. 2. Methods A conceptual model was developed to investigate the relationships between factors affecting consumers’ attitude towards purchasing wood products originating from transgenic plantations and consumer groups, who exhibit a specific buying behaviour. The model also explored the linkage between consumers, who exhibit a specific buying behaviour and their opinion towards establishment of transgenic forest plantations (Figure 1). Based on the conceptual model the null research hypotheses tested in this study were the following: • Ho1: Consumers can not be classified into groups according to their purchasing behaviour towards wood products originating from transgenic plantations. • Ho2: Consumers’ opinion towards establishment of transgenic forest plantations is not significant in relation to a particular buying behaviour.
555
556
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560
H o2
Ho1 Factors affecting consumers’ buying behaviour towards transgenic wood products
Consumers’ groups according to their buying behaviour
Consumers’ opinion towards establishment of transgenic plantations
Figure 1. Conceptual Model of consumers’ attitudes towards transgenic wood products
A survey with face-to face interviews throughout the country was undertaken to gather the necessary information. Random systematic sampling was used to form the sample by taking every sixth person that came on site [1, 13, 12]. A total of 418 consumers were included in the sample. Statistical analysis however was performed on 231 consumers (55.26%), those who declared that they would be willing to buy wood products originating from genetically modified forest trees. The sample comprised 0.003% of the total population and it was considered reasonably representative of it, as all the sample characteristics did not differ from those of the total population, based on Census data [4, 18]. The questionnaire designed to meet the research objectives was piloted in October 2011 to 30 consumers. The pilot survey indicated that it needed no modification and therefore, the main survey was conducted in November and December 2011. Multivariate analysis techniques were applied in three stages to the responses of the 231 consumers to reveal the key information these contained. Principal Component Analysis (PCA) was used to identify the variables that accounted for the maximum amount of variance within the data, in terms of the smallest number of uncorrelated variables (components). The anti-image correlation matrix, as well as the Bartlett’s test of sphericity and the Measure of Sampling Adequacy (MSA) were used, in order to check the appropriateness of the data for subsequent factor analysis. Variables with a high proportion of large absolute values of anti –image correlations, as well as MSA less than 0.5, were removed before analysis. An orthogonal rotation (varimax method) was conducted and the standard criteria of eigenvalue = 1, scree test and percentage of variance were used in order to determine the factors in the first rotation [10]. Different trial rotations followed, where factor interpretability was used to compare the reduced 7 PCA variables, which related consumers’ purchase behaviour towards transgenic wood products to a smaller set of underlying factors. These PCA scores were then subjected to cluster analysis to group consumers with similar patterns of scores into similar clusters of buying behaviour. Both hierarchical and non-hierarchical methods were used [10], in order to develop a typology of the consumers’ buying behaviour. Cluster analysis was conducted on all 231 observations, as there were no outliers. Quadratic Discriminant Analysis was performed to assess how accurately the key factors identified through factor analysis could predict and discriminate cluster membership. Furthermore, the Friedman one way test was performed to identify the relationship between the consumers’ opinion towards establishment of transgenic forest plantations and their particular buying behaviour. 3. Results The factors affecting consumers’ purchasing behaviour towards transgenic wood products along with their eigenvalues, the scree plot test and the percentage of variance results from PCA and Factor Analysis are portrayed in Table 1 and the three main key factors in Table 2.
557
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 1. Results of Principal Components Analysis regarding identification of variables affecting consumers’ buying behaviour towards transgenic wood products Component Eigenvalues % of Variance Cumulative % of Variables Communalities Variance 1 2.130 30.431 30.431 Quality 0.647 2 1.469 20.990 51.421 Brand name 0.582 3 1.065 15.215 66.636 Special Characteristics 0.599 4 0.660 9.434 76.070 Certification of origin 0.723 5 0.650 9.283 85.393 Labelled as products made of 0.766 wood derived from GM plantations 6 0.548 7.830 93.184 Attractiveness of packing 0.677 7 0.477 6.816 100.00 Advertisement 0.671 Table 2. Key Factors affecting consumers’ purchasing behaviour towards transgenic wood products KEY ATTITUDE DIMENSIONS
Factor Loading
Promotion Issues Attractiveness of packing Advertisement
0.819 0.807 Product Features
Quality Special Characteristics Brand name
0.802 0.719 0.614 Labelling
Labelled as products made of wood derived from GM plantations Certification of origin
0.873 0.811
KMO MSA = 0.638, Bartlett test of Sphericity = 223.017, P <0.001
Three groups of consumers, named according to their buying behaviour towards transgenic wood products were formed (Table 3): (a) Consumers influenced by marketing issues, (b) Opportunists (c) Consumers interested in product features. Table 3. Classification of consumers according to their buying behaviour Factors Groups of Consumers
Promotion Issues Product Features Labelling Number of Consumers ( N =231)
Consumers influenced by marketing issues 0.33152 -0.97946 0.59869 69
Opportunists
P Consumers interested in product features
-1.07330 -0.4886 -0.4994 70
0.56800 0.77177 -0.6862 92
0.001 0.001 0.001
Consumers, who are influenced by marketing issues, comprise 30% of the potential buyers of transgenic wood products of the sample. These consumers would buy transgenic wood products, but they would like the products to be labelled as products derived from GM plantations. They are also interested in the origin of the products, the attractiveness of packing and their advertisement. On the other hand, they are not interested in product features. The opportunists comprise also 30% of the potential buyers of transgenic wood products of the sample. These consumers would buy transgenic wood products without being influenced by any particular factor in their choice. Consumers, who are interested in product features, comprise 40% of the potential buyers of transgenic wood products of the sample. These consumers are influenced in their purchasing decision by the quality of the product, its special characteristics and its brand name. They also interested in the attractiveness of its packing and its advertisement, while they are not interested in labelling issues. A summary of the cross validation classification derived by the quadratic Discriminant Analysis is shown in Table 4.
558
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 4. Prediction of consumers’ group membership with cross - validation Actual Classification Opportunists Consumers interested in product features Consumers interested in marketing issues Total N N Correct Proportion of Correct Classification
Consumers influenced by marketing issues 68 1 0 69 68 98.6%
Predicted Classification Opportunists Consumers interested in product features 0 70 0 70 70 100%
2 2 88 92 88 95.7%
It is evident that, the three attitude dimensions could accurately predict and discriminate consumers’ group membership. Therefore, the hypothesis Ho1: Consumers cannot be classified into groups according to their purchasing behaviour towards wood products originating from transgenic plantations could be rejected. The Friedman non- parametric test was employed to explore consumers’ opinion towards establishment of transgenic plantations (Table 5). Table 5 indicates that most of the consumers have a similar opinion towards establishment of transgenic forest plantations. In particular, most of the consumers whose purchase decision towards transgenic wood products is influenced by marketing issues believe that establishment of forest transgenic plantations might contribute to the increase of job flexibility, reduction of production cost, increase of farmers’ income and reduction of production losses. On the other hand, the opportunists think that establishment of transgenic forest plantations will reduce the production costs. They also believe that adoption of GM forest plantations will contribute to an increase of job flexibility, increase of income and reduction of production losses. The consumers, who are interested in product features, consider that the adoption of GM forest plantations will lead to an increase of job flexibility, reduction of production cost and production losses, as well as to an increase of the farmers’ income. These findings comply with similar issues reported by other studies [14, 15, 5, 18, 3, 2, 9, 6, 11]. The environmental issues that may arise from the potential adoption of transgenic forest plantations do not appear so important to all consumers’ groups in Greece, despite such issues have been stressed in several studies [7,15, 6]. Hence the hypothesis, Ho2: Consumers’ opinion towards establishment of transgenic forest plantations is not significant in relation to a particular buying behaviour could be rejected.
559
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 5. Prediction of consumers’ group membership with cross - validation Consumers' Opinion about the Use of Consumers influenced by Biotechnology in Forest tree sector marketing issues (x2=20,677, df = 10, P<0,023)
Opportunists (x2=20,677, df = 10, P<0,001)
Consumers interested in product features (x2=61,845, df = 10, P<0,001)
The use of biotechnology increase of job flexibility The use of biotechnology reduce Cost Production The use of biotechnology increase Income
7.23
6.9
7.46
6.67
7.01
6.84
6.23
6.75
6.59
The use of biotechnology reduce the production losses The use of biotechnology has negative environmental of impact The use of biotehcnology has negative impact on biodiversity The use of biotechnology has negative impact on non GMO products The use of biotechnology may harm human health Biotechnology is considered not important
5.64
6.6
6.65
5.38
5.45
5.08
5.79
5.69
5.68
6.02
5.64
5.58
5.99
5.46
5.41
5.67
4.7
5.2
The use of biotechnology hase negative impact on climate change The use of biotechnology is important for biomas production
5.73
5.32
5.3
5.66
6.49
6.21
4. Conclusions From the results of this study, it appears that there might be potential buyers of wood products originating from transgenic plantations in Greece. Indeed, most of these potential consumers willing to purchase transgenic wood products base their buying decisions on economic issues. Taking into consideration that most of these products are not directly linked with human health impacts, there is a potential for development of a market for such products, particularly in current times of economic depression. Hence, the potential market developers should structure their marketing and promotion mix and focus on quality of product and efficiency of production methods, as well as on effective advertisement, attractive packing and labelling issues. On the other hand, the empirical results of this study, which to our knowledge is the first of its kind in Europe, should trigger further studies both in Greece and other parts of Europe, so as to gradually build the required scientific knowledge base that in combination with environmental risk assessments and biosafety standards will help policy makers at national and EU level, to more adequately shape regulatory policies for commercialization of transgenic forest plantations. References [1] Barnett, V. Sample survey, principles and methods. Edward Arnold, Kent; 1991, p. 1-173. [2] Brunner, A.M., DiFazio, J.Li, S.P.,Shevchenko, O., Montgomery, B.E., Mohamed, R., Wei, H., Ma, C., Elias, A.A., Vanwormer, K. and Strauss, S.H.., Genetic containment of forest plantations. Tree Genetics & Genomes; 2007;3:75-100. [3] Carman, N., Langelle, O., Perry, A., Petermann, A., Smith, D., J.D., Tokar, B. Ecological and Social Impacts of East Growing Timber Plantations and Genetically Engineered Trees, http://www.forestethics.org/downloads/GEtreeport.pdf; 2006. [4] Chen, M. Consumers attitudes and purchase intentions in relation to organic foods in Taiwan: Moderating effects of food-related personality traits. Food Quality and Preference; 2007;18:1008-1021. [5] El-Lakany, M.H. Are genetically modified trees a threat to forests? Unasylva 2004;217, Vol.55, No.1:45-47. [6] FAO. The potential environmental, cultural and socio-economic impacts of genetically modified trees. UNEP/CBD/SBSTTA/13/INF/6; 2008; pp17. [7] Gartland, K., Crow R., Fenning T. and Gartland J., Genetically Modified Trees: Production, Properties, and Potential, Journal of Arboriculture; 2003;29 (5):259-266. [8] Häggman, H., Find, J.M., Pilate, G., Gallardo, F., Ruohonen-Lehto, M., Kazana V., Migliacci, F., Ionita, L., Sijacic-Nikolic, M., Donnarumma, F., Harfouche, A., Biricolti, S., Glandorf, B., Tsourgiannis, L., Minol, K., Paffetti, D., Fladung, M. and Vettori, C. Biosafety
560
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 of genetically modified forest trees (GMTs) –COST Action FP0905- a common action of European scientists. 2nd International Conference of the IUFRO Working Party 2.09.02. Mendel lectures & Plenary MLP-3; 2012, p.13. [9] Halpin, C., Thain, S.C., Tilston, E.L., Guiney, E., Lapierre, C. and Hopkins, D.W. Ecological impacts of trees with modified lignin, Tree Genetics & Genomes, 2007;3(2):101-110. [10] Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. Multivariate data analysis. Prentice Hall Inc, New Jersey; 1998. [11] Harfouche, A., Meilan, R. and Altman, A.Tree genetic engineering and applications to sustainable forestry and biomass production. Trends in Biotechnology, 2011;Vol. 29, No.: p. 11-17. [12] McCluskey, J., Grimsrud, K., Ouchi, H, and Wahl, T. Consumer Response to Genetically Modified Food Products in Japan, Agricultural and Resource Economic Review, 2003;32 (2):222-231. [13] Oppenheim, A. N. Questionnaire design, interviewing and attitude measurement. Continuum, New York; 2000. [14] Serjo, R.A. Transgenic Trees: Implementation and Outcomes of the Plant Protection Act, April 2004, Resources for the Future. [15] Sedjo, R.A. Toward commercialization of genetically engineered forests: economic and social considerations. Resources for the Future,2006; pp 46. [16] Sedjo, R. Transgenic trees for biomass: The effects of regulatory restrictions and court decisions on the pace of commercialization. AgBioForum, 2010;13(4):391-397. [17] Thomas S. Ethical and social considerations in commercial uses of food and fibber crops, Proceedings of the First International Symposium on Ecological and Societal Aspects of Transgenic Plantations, 2001;92-98. [18] Tsourgiannis, L., Eddison, J. and Warren, M. Factors affecting the marketing channel choice of sheep and goat farmers in the region of East Macedonia in Greece regarding the distribution of their milk production. Small Ruminant Research, 2008;79:87-97 [18] Van Frankenhuyzen K., and Beardmore, T., Current status and environmental impact of transgenic forest trees, Canadian Journal of Forest Research, 2004;34:1163-1180.
ScienceDirect Procedia Technology 8 (2013) 554 – 560
6th International Conference on Information and Communication Technologies in Agriculture, Food and Environment (HAICTA 2013)
Exploring Consumers’ Attitudes Towards Wood Products that could be Derived from Transgenic Plantations in Greece Lambros Tsourgiannisa,b*, Vasiliki Kazanab, Anastasios Karasavvoglouc, Michael Nikolaidisc, Giannoula Florouc, Persefoni Polychronidouc b
a Region of East Macedonia and Thrace, Regional District of Xanthi, 67100 Xanthi, Greece Department of Forestry and Natural Environment Management at Drama, Kavala Institute of Technology , Drama, Greece c Department of Accountancy, School of Business and Economy, Kavala Institute of Technology, 65404, Kavala, Greece
Abstract This paper aims at exploring consumers’ purchasing behaviour towards wood products that could be derived from transgenic plantations. Field interviews were conducted and analysis was performed on a sample of 231 consumers, who declared that they would buy such wood products. Principal Components Analysis indicated that the main factors affecting consumers’ purchasing behaviour towards those products are promotion issues, product features and labelling. Through cluster techniques 3 groups of consumers with similar consumption behaviour were identified: ((a) Consumers influenced by marketing issues, (b) Opportunists (c) Consumers interested in product features. Discriminant Analysis was performed to predict cluster membership and a non parametric test to profile each consumer group according to personal characteristics and attitude towards establishment of transgenic forest plantations. © 2013 byby Elsevier Ltd.B.V. © 2014 The TheAuthors. Authors.Published Published Elsevier Selection and under responsibility of The Hellenic Association for Information and Communication Technologies in Agriculture Selection andpeer-review peer-review under responsibility of HAICTA. Food and Environment (HAICTA)
Keywords: Consumer behaviour; wood products; Forest Transgenic Trees.
* Corresponding author. Tel.: +30 6972084118. E-mail address: [email protected]
2212-0173 © 2013 The Authors. Published by Elsevier Ltd. Selection and peer-review under responsibility of The Hellenic Association for Information and Communication Technologies in Agriculture Food and Environment (HAICTA) doi:10.1016/j.protcy.2013.11.078
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560
1. Introduction Transgenic forest trees are biotechnology products, that is, they are organisms Genetically Modified (GM) or Genetically Engineered (GE) through the insertion of DNA from other species in order to manifest specific traits (www.forestguild.org). Such traits mainly concern higher growth rates, lignin reduction and increased resistance to herbicides or forest pests. At the current state of art there are five main domains of potential application of GM trees: basic and applied research, commercial production, bio-remediation, ecological restoration and response to epidemics. Much discussion has been geared by academics and the industry on the potential of GM forest trees to help meet projected demand increases for wood products worldwide, through increased growth and productivity and decreased cost of wood fiber production [5, 18, 3, 16, 2, 9, 6, 11]. However, with the exception of China, no market for GM forest trees currently exists in Europe, the USA or other parts of the world, despite that over the past 20 years through confined transgenic forest tree field trials and millions of hectares of commercial transgenic crop fields much scientific experience has been gained [8]. The main reasons for the absence of such markets are the very strict regulatory frameworks regarding license granting for establishment of GM plantations, particularly due to a number of potential public concerns. These include the potential for spread of antibiotic or herbicide resistance genes to other non-target species from GM trees; the potential for long – distance pollen spread over many years from long – lived trees, the potential for adverse effects on biodiversity from transgenic forest plantations and any unexpected effects [7, 15, 6]. Also, GM forest plantations might generate profit for certain actors in the private sector, while poorer communities will be further marginalized [17]. The regulation framework of transgenic forest trees in use differs between countries. In the European Union regulations aim at providing high levels of protection for human health and the environment, while ensuring consumer interests and the promotion of competitive market. Socio-economic information of transgenic forest plantations in Europe is completely lacking. Evidence from other countries, such as the USA, indicated that the industrial consumers of wood products, who use wood as an input to production, are generally enthusiastic about transgenic forest trees, contrary to the attitude of consumers towards final wood products, who appear more concerned [14]. Although wood products from transgenic plantations are unlikely to be in the market for another 10-15 years, scientifically based information on the anticipated attitudes of consumers is extremely important both for developers as without the expectation of viable market the developments and investments are unlikely to be forthcoming and for policy makers, who need to respond adequately through regulation tools and programs. Hence, this paper presents results of a study, the first of its kind in Europe to the authors’ knowledge, initiated in Greece to explore the attitudes of consumers towards wood products that could be derived from transgenic forest plantations and profile them according to personal characteristics and their attitude towards establishment of such plantations. 2. Methods A conceptual model was developed to investigate the relationships between factors affecting consumers’ attitude towards purchasing wood products originating from transgenic plantations and consumer groups, who exhibit a specific buying behaviour. The model also explored the linkage between consumers, who exhibit a specific buying behaviour and their opinion towards establishment of transgenic forest plantations (Figure 1). Based on the conceptual model the null research hypotheses tested in this study were the following: • Ho1: Consumers can not be classified into groups according to their purchasing behaviour towards wood products originating from transgenic plantations. • Ho2: Consumers’ opinion towards establishment of transgenic forest plantations is not significant in relation to a particular buying behaviour.
555
556
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560
H o2
Ho1 Factors affecting consumers’ buying behaviour towards transgenic wood products
Consumers’ groups according to their buying behaviour
Consumers’ opinion towards establishment of transgenic plantations
Figure 1. Conceptual Model of consumers’ attitudes towards transgenic wood products
A survey with face-to face interviews throughout the country was undertaken to gather the necessary information. Random systematic sampling was used to form the sample by taking every sixth person that came on site [1, 13, 12]. A total of 418 consumers were included in the sample. Statistical analysis however was performed on 231 consumers (55.26%), those who declared that they would be willing to buy wood products originating from genetically modified forest trees. The sample comprised 0.003% of the total population and it was considered reasonably representative of it, as all the sample characteristics did not differ from those of the total population, based on Census data [4, 18]. The questionnaire designed to meet the research objectives was piloted in October 2011 to 30 consumers. The pilot survey indicated that it needed no modification and therefore, the main survey was conducted in November and December 2011. Multivariate analysis techniques were applied in three stages to the responses of the 231 consumers to reveal the key information these contained. Principal Component Analysis (PCA) was used to identify the variables that accounted for the maximum amount of variance within the data, in terms of the smallest number of uncorrelated variables (components). The anti-image correlation matrix, as well as the Bartlett’s test of sphericity and the Measure of Sampling Adequacy (MSA) were used, in order to check the appropriateness of the data for subsequent factor analysis. Variables with a high proportion of large absolute values of anti –image correlations, as well as MSA less than 0.5, were removed before analysis. An orthogonal rotation (varimax method) was conducted and the standard criteria of eigenvalue = 1, scree test and percentage of variance were used in order to determine the factors in the first rotation [10]. Different trial rotations followed, where factor interpretability was used to compare the reduced 7 PCA variables, which related consumers’ purchase behaviour towards transgenic wood products to a smaller set of underlying factors. These PCA scores were then subjected to cluster analysis to group consumers with similar patterns of scores into similar clusters of buying behaviour. Both hierarchical and non-hierarchical methods were used [10], in order to develop a typology of the consumers’ buying behaviour. Cluster analysis was conducted on all 231 observations, as there were no outliers. Quadratic Discriminant Analysis was performed to assess how accurately the key factors identified through factor analysis could predict and discriminate cluster membership. Furthermore, the Friedman one way test was performed to identify the relationship between the consumers’ opinion towards establishment of transgenic forest plantations and their particular buying behaviour. 3. Results The factors affecting consumers’ purchasing behaviour towards transgenic wood products along with their eigenvalues, the scree plot test and the percentage of variance results from PCA and Factor Analysis are portrayed in Table 1 and the three main key factors in Table 2.
557
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 1. Results of Principal Components Analysis regarding identification of variables affecting consumers’ buying behaviour towards transgenic wood products Component Eigenvalues % of Variance Cumulative % of Variables Communalities Variance 1 2.130 30.431 30.431 Quality 0.647 2 1.469 20.990 51.421 Brand name 0.582 3 1.065 15.215 66.636 Special Characteristics 0.599 4 0.660 9.434 76.070 Certification of origin 0.723 5 0.650 9.283 85.393 Labelled as products made of 0.766 wood derived from GM plantations 6 0.548 7.830 93.184 Attractiveness of packing 0.677 7 0.477 6.816 100.00 Advertisement 0.671 Table 2. Key Factors affecting consumers’ purchasing behaviour towards transgenic wood products KEY ATTITUDE DIMENSIONS
Factor Loading
Promotion Issues Attractiveness of packing Advertisement
0.819 0.807 Product Features
Quality Special Characteristics Brand name
0.802 0.719 0.614 Labelling
Labelled as products made of wood derived from GM plantations Certification of origin
0.873 0.811
KMO MSA = 0.638, Bartlett test of Sphericity = 223.017, P <0.001
Three groups of consumers, named according to their buying behaviour towards transgenic wood products were formed (Table 3): (a) Consumers influenced by marketing issues, (b) Opportunists (c) Consumers interested in product features. Table 3. Classification of consumers according to their buying behaviour Factors Groups of Consumers
Promotion Issues Product Features Labelling Number of Consumers ( N =231)
Consumers influenced by marketing issues 0.33152 -0.97946 0.59869 69
Opportunists
P Consumers interested in product features
-1.07330 -0.4886 -0.4994 70
0.56800 0.77177 -0.6862 92
0.001 0.001 0.001
Consumers, who are influenced by marketing issues, comprise 30% of the potential buyers of transgenic wood products of the sample. These consumers would buy transgenic wood products, but they would like the products to be labelled as products derived from GM plantations. They are also interested in the origin of the products, the attractiveness of packing and their advertisement. On the other hand, they are not interested in product features. The opportunists comprise also 30% of the potential buyers of transgenic wood products of the sample. These consumers would buy transgenic wood products without being influenced by any particular factor in their choice. Consumers, who are interested in product features, comprise 40% of the potential buyers of transgenic wood products of the sample. These consumers are influenced in their purchasing decision by the quality of the product, its special characteristics and its brand name. They also interested in the attractiveness of its packing and its advertisement, while they are not interested in labelling issues. A summary of the cross validation classification derived by the quadratic Discriminant Analysis is shown in Table 4.
558
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 4. Prediction of consumers’ group membership with cross - validation Actual Classification Opportunists Consumers interested in product features Consumers interested in marketing issues Total N N Correct Proportion of Correct Classification
Consumers influenced by marketing issues 68 1 0 69 68 98.6%
Predicted Classification Opportunists Consumers interested in product features 0 70 0 70 70 100%
2 2 88 92 88 95.7%
It is evident that, the three attitude dimensions could accurately predict and discriminate consumers’ group membership. Therefore, the hypothesis Ho1: Consumers cannot be classified into groups according to their purchasing behaviour towards wood products originating from transgenic plantations could be rejected. The Friedman non- parametric test was employed to explore consumers’ opinion towards establishment of transgenic plantations (Table 5). Table 5 indicates that most of the consumers have a similar opinion towards establishment of transgenic forest plantations. In particular, most of the consumers whose purchase decision towards transgenic wood products is influenced by marketing issues believe that establishment of forest transgenic plantations might contribute to the increase of job flexibility, reduction of production cost, increase of farmers’ income and reduction of production losses. On the other hand, the opportunists think that establishment of transgenic forest plantations will reduce the production costs. They also believe that adoption of GM forest plantations will contribute to an increase of job flexibility, increase of income and reduction of production losses. The consumers, who are interested in product features, consider that the adoption of GM forest plantations will lead to an increase of job flexibility, reduction of production cost and production losses, as well as to an increase of the farmers’ income. These findings comply with similar issues reported by other studies [14, 15, 5, 18, 3, 2, 9, 6, 11]. The environmental issues that may arise from the potential adoption of transgenic forest plantations do not appear so important to all consumers’ groups in Greece, despite such issues have been stressed in several studies [7,15, 6]. Hence the hypothesis, Ho2: Consumers’ opinion towards establishment of transgenic forest plantations is not significant in relation to a particular buying behaviour could be rejected.
559
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 Table 5. Prediction of consumers’ group membership with cross - validation Consumers' Opinion about the Use of Consumers influenced by Biotechnology in Forest tree sector marketing issues (x2=20,677, df = 10, P<0,023)
Opportunists (x2=20,677, df = 10, P<0,001)
Consumers interested in product features (x2=61,845, df = 10, P<0,001)
The use of biotechnology increase of job flexibility The use of biotechnology reduce Cost Production The use of biotechnology increase Income
7.23
6.9
7.46
6.67
7.01
6.84
6.23
6.75
6.59
The use of biotechnology reduce the production losses The use of biotechnology has negative environmental of impact The use of biotehcnology has negative impact on biodiversity The use of biotechnology has negative impact on non GMO products The use of biotechnology may harm human health Biotechnology is considered not important
5.64
6.6
6.65
5.38
5.45
5.08
5.79
5.69
5.68
6.02
5.64
5.58
5.99
5.46
5.41
5.67
4.7
5.2
The use of biotechnology hase negative impact on climate change The use of biotechnology is important for biomas production
5.73
5.32
5.3
5.66
6.49
6.21
4. Conclusions From the results of this study, it appears that there might be potential buyers of wood products originating from transgenic plantations in Greece. Indeed, most of these potential consumers willing to purchase transgenic wood products base their buying decisions on economic issues. Taking into consideration that most of these products are not directly linked with human health impacts, there is a potential for development of a market for such products, particularly in current times of economic depression. Hence, the potential market developers should structure their marketing and promotion mix and focus on quality of product and efficiency of production methods, as well as on effective advertisement, attractive packing and labelling issues. On the other hand, the empirical results of this study, which to our knowledge is the first of its kind in Europe, should trigger further studies both in Greece and other parts of Europe, so as to gradually build the required scientific knowledge base that in combination with environmental risk assessments and biosafety standards will help policy makers at national and EU level, to more adequately shape regulatory policies for commercialization of transgenic forest plantations. References [1] Barnett, V. Sample survey, principles and methods. Edward Arnold, Kent; 1991, p. 1-173. [2] Brunner, A.M., DiFazio, J.Li, S.P.,Shevchenko, O., Montgomery, B.E., Mohamed, R., Wei, H., Ma, C., Elias, A.A., Vanwormer, K. and Strauss, S.H.., Genetic containment of forest plantations. Tree Genetics & Genomes; 2007;3:75-100. [3] Carman, N., Langelle, O., Perry, A., Petermann, A., Smith, D., J.D., Tokar, B. Ecological and Social Impacts of East Growing Timber Plantations and Genetically Engineered Trees, http://www.forestethics.org/downloads/GEtreeport.pdf; 2006. [4] Chen, M. Consumers attitudes and purchase intentions in relation to organic foods in Taiwan: Moderating effects of food-related personality traits. Food Quality and Preference; 2007;18:1008-1021. [5] El-Lakany, M.H. Are genetically modified trees a threat to forests? Unasylva 2004;217, Vol.55, No.1:45-47. [6] FAO. The potential environmental, cultural and socio-economic impacts of genetically modified trees. UNEP/CBD/SBSTTA/13/INF/6; 2008; pp17. [7] Gartland, K., Crow R., Fenning T. and Gartland J., Genetically Modified Trees: Production, Properties, and Potential, Journal of Arboriculture; 2003;29 (5):259-266. [8] Häggman, H., Find, J.M., Pilate, G., Gallardo, F., Ruohonen-Lehto, M., Kazana V., Migliacci, F., Ionita, L., Sijacic-Nikolic, M., Donnarumma, F., Harfouche, A., Biricolti, S., Glandorf, B., Tsourgiannis, L., Minol, K., Paffetti, D., Fladung, M. and Vettori, C. Biosafety
560
Lambros Tsourgiannis et al. / Procedia Technology 8 (2013) 554 – 560 of genetically modified forest trees (GMTs) –COST Action FP0905- a common action of European scientists. 2nd International Conference of the IUFRO Working Party 2.09.02. Mendel lectures & Plenary MLP-3; 2012, p.13. [9] Halpin, C., Thain, S.C., Tilston, E.L., Guiney, E., Lapierre, C. and Hopkins, D.W. Ecological impacts of trees with modified lignin, Tree Genetics & Genomes, 2007;3(2):101-110. [10] Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. Multivariate data analysis. Prentice Hall Inc, New Jersey; 1998. [11] Harfouche, A., Meilan, R. and Altman, A.Tree genetic engineering and applications to sustainable forestry and biomass production. Trends in Biotechnology, 2011;Vol. 29, No.: p. 11-17. [12] McCluskey, J., Grimsrud, K., Ouchi, H, and Wahl, T. Consumer Response to Genetically Modified Food Products in Japan, Agricultural and Resource Economic Review, 2003;32 (2):222-231. [13] Oppenheim, A. N. Questionnaire design, interviewing and attitude measurement. Continuum, New York; 2000. [14] Serjo, R.A. Transgenic Trees: Implementation and Outcomes of the Plant Protection Act, April 2004, Resources for the Future. [15] Sedjo, R.A. Toward commercialization of genetically engineered forests: economic and social considerations. Resources for the Future,2006; pp 46. [16] Sedjo, R. Transgenic trees for biomass: The effects of regulatory restrictions and court decisions on the pace of commercialization. AgBioForum, 2010;13(4):391-397. [17] Thomas S. Ethical and social considerations in commercial uses of food and fibber crops, Proceedings of the First International Symposium on Ecological and Societal Aspects of Transgenic Plantations, 2001;92-98. [18] Tsourgiannis, L., Eddison, J. and Warren, M. Factors affecting the marketing channel choice of sheep and goat farmers in the region of East Macedonia in Greece regarding the distribution of their milk production. Small Ruminant Research, 2008;79:87-97 [18] Van Frankenhuyzen K., and Beardmore, T., Current status and environmental impact of transgenic forest trees, Canadian Journal of Forest Research, 2004;34:1163-1180.