Using involvement to understand individual responses to an issue: the case of New Zealand biosecurity

Abstract

The effectiveness of biosecurity measures at national borders is influenced by the behaviour and levels of involvement of travellers. Involvement is the importance or relevance of an object or situation to an individual. Involvement helps regulate the way in which people receive and process information and thus influences the extent of information searching for decision making, and information processing and persuasion. This work drew on the concept of involvement to investigate the response of individuals to New Zealand biosecurity requirements. A convenience sample of people associated with the agricultural and food processing sectors were surveyed using a five-item scale of involvement to measure their level of involvement in biosecurity. The results indicated that most respondents had medium to high levels of involvement. This implies that respondents were motivated to attend to and process information on biosecurity measures. However, not all respondents reported taking note of biosecurity information, implying that involvement with biosecurity prompts some initial information processing that may or may not continue over time.

Keywords:

• behaviour
• involvement
• biosecurity
• travellers
• information processing

Introduction

New Zealand's economy depends heavily on agriculture and tourism (Goldson et al. 2005; Jay et al. 2003). Historically, New Zealand's geographic isolation helped exclude unwanted exotic species from entering the country, but increased trade and travel has resulted in a corresponding increase in the risk of biosecurity incursions (Goldson et al. 2005; Hall 2005; Jay et al. 2003; Kriticos et al. 2005). The impact of human-mediated accidental and deliberate introduction of exotic species is considerable (Andreu et al. 2009; Brasier 2008; Vitousek et al. 1997). Kriticos et al. (2005) estimated that, with no improvements to the biosecurity system, New Zealand would have to deal with 542 potential pest incursions between 2005 and 2017. Taking into account direct impacts and ongoing control costs, this would cost the economy NZ$921 million (Kriticos et al. 2005). Further, Kriticos et al. (2005) estimated that improving the rate of detection and interception of exotic species at the border by 10% would reduce expenditure on incursions by $16 million over the same time period.

A range of measures is used to reduce the number of biosecurity incursions. These include pre-border measures such as checking and treating imported goods in the country of origin, border control measures such as screening and inspection, post-border surveillance programmes (Jay et al. 2003; Kean et al. 2008) and eradication responses (Kean & Suckling 2005). People are an integral part of the biosecurity system. There are numerous interactions between people at an individual and organisational level and at different points in the system from the pre-border through to the post-border environment. Their behaviour, in response to the measures outlined above, will determine whether biosecurity risks can be easily managed. Their response to a particular measure will depend on a range of factors, including their perception of biosecurity. Therefore, understanding people's perception of biosecurity is critical to effectively managing the risk of biosecurity incursions (García-Llorente et al. 2008).

The New Zealand biosecurity system

New Zealand's biosecurity system is considered to be one of the most comprehensive in the world (Loope 2004; Meyerson & Reaser 2002). The current biosecurity system in New Zealand consists of a Biosecurity Act, introduced in 1993; a Biosecurity Strategy, released in 2003 and endorsed by government; and a lead government agency, the Ministry of Agriculture and Forestry (MAF). The Biosecurity Act, passed by the New Zealand government in 1993, was a means of amalgamating several different Acts all relating to biosecurity (Webb 1995). The stated purpose of the Act is to eradicate and manage unwanted organisms already in the country and to prevent unwanted organisms from entering (Storey & Clayton 2002). The Act outlines the roles of importers, landholders and MAF (Webb 1995). The Act also specifies that the public have a duty to report notifiable organisms (Webb 1995). Travellers to New Zealand are also a biosecurity risk (Forer & McNeill 2008; Young et al. 2007) and are required to declare any biosecurity risk items before crossing the border.

Understanding response to regulations

Traditionally, most approaches to understanding individual behaviour towards regulations, such as biosecurity requirements at the border, have been based on the rational choice model of deterrence theory (Frank 1987; Winter & May 2001). The rational choice model assumes individuals will pursue self-interest and that this is the primary motivator for their behaviour (Akers 1990). Under these circumstances, an individual's perception of the likelihood of detection and the penalty involved determines how effective enforcement strategies are (Kaine et al. 2010; Winter & May 2001). However, there are numerous examples where this does not occur (i.e. individuals choose not to follow the rules), even if it is not in their self-interest to do so (Frank 1987; Sutinen & Kuperan 1999). Some authors have advocated combining economic, sociological and psychological approaches from the literature to help understand human behaviour in these circumstances (Sutinen & Kuperan 1999; Winter & May 2001).

There are a number of human behaviour models available from sociology and psychology (Fishbein & Ajzen 1975; Petty et al. 1983; Weinstein & Sandman 1992). Most recognise the fundamental role that attitudes appear to have on human behaviour (Kaine et al. 2010). The main limitation of these models is the assumption that a decision is sufficiently important to an individual for them to go to the effort of forming an attitude (Kaine et al. 2010; Priluck & Till 2004). Attitudes that align strongly to behaviour tend to be strongly held and easily accessible (Fazio & Olsen 2003; Glasman & Albarracín 2006; Kokkinaki & Lunt 1999).

Involvement, defined as a measure of the intensity of an individual's motivation in regard to a decision (Verbeke & Vackier 2004), helps put some of the models of human behaviour mentioned above into context. If a person is highly involved they will put time and effort into forming an attitude. If they are not involved, while they may be able to express an attitude, it may not be particularly reflective of their behaviour (Kaine et al. 2010). Understanding an individual's involvement in an issue should provide some information on how strongly their attitudes are held and whether their attitudes will be a good predictor of their behaviour.

Describing involvement

Involvement is viewed as a function of the person, object and situation, and has a strength or intensity determined by the mobilisation of resources in order to achieve a goal (Poiesz & deBont 1995). The strength of involvement is determined by the relevance of a product (or activity) to an individual's values, goals and self-concept (Zaichkowsky 1986).

Involvement has been shown to regulate the way in which people receive and process information (Heath & Douglas 1991; Salmon 1986). Audience involvement is therefore perceived to be important for designers of information and communication campaigns because involvement will provide some indication of how audiences may approach the information provided (Gregory 2004; Kim 2003). In effect, ‘Involvement determines whether the audience is active or passive …’ (Roser 1990, p. 571). Communication campaigns tend to be most effective among people who are involved in the issue already (Gregory 2004; Larson & Massetti-Miller 1984). Audiences with low involvement are the most challenging (Gregory 2004).

Generally, those people who are highly involved with an issue will be prepared to devote time and effort to evaluating information associated with that issue and will already have appropriate frames of reference (Heath & Douglas 1991; Petty et al. 1983; Rimal & Real 2005). However, those who have low involvement may notice other cues, such as the way in which the information is presented (Rimal & Real 2005) and the message source (Petty et al. 1983). This can lead to behaviour that is defined by a specific situation and does not reflect an individual's attitude or knowledge (Chaffee & Roser 1986).

Involvement has been used to understand individuals’ behaviour in a range of contexts. Recent research on involvement has explored involvement in fashion (O'Cass 2000), food (Pieniak et al. 2008; Verbeke 2008), leisure activities (Havitz & Dimanche 1997, 1999; Kyle et al. 2007), health (Gregory 2004, 2006; Kim 2006) and services (Bienstock & Stafford 2006). Involvement has also been used by health professionals to help design communication campaigns to promote behaviour change (Aldoory 2001). In agriculture, Kaine (2008) used involvement as the basis for understanding and predicting the adoption of technologies.

Methods

Measuring involvement

Many scales have been developed to measure involvement. O'Cass (2000) found that 23 measures had been developed to measure involvement in the last 40 years. These ranged from simple elicitation of overall level of involvement (Zaichkowsky 1985) to measuring involvement across several dimensions, thus identifying source of involvement (Kapferer & Laurent 1985). A review of the literature revealed a number of potential involvement scales. One was chosen to measure involvement in biosecurity: Mittal's (1995) modification of Zaichkowsky's (1985) involvement scale, the Personal Involvement Inventory (PII).

Zaichkowsky's (1985) involvement scale, the PII, was designed to measure involvement defined as ‘a person's perceived relevance of the object based on inherent needs, values and interests’ (Zaichkowsky 1985, p. 342). The PII covered personal, physical and situational involvement to provide an overall measure of involvement. It was a simple scale with 20 word pairs used to represent different aspects of involvement such as ‘important/unimportant’ and ‘of no concern/of concern to me’. Each item was added up to give a total score of involvement between 20 and 140. McQuarrie & Munson (1987) revised Zaichkowsky's PII, renaming it RPII. They reduced the scale to 14 word pairs and incorporated some of Kapferer & Laurent's (1985) items into the scale in an attempt to account for different dimensions of involvement. The RPII was tested on 12 products. Later, McQuarrie & Munson (1992) revised it again, reducing it to 10 items. Mittal (1995) further refined the PII, reducing it to five items. Mittal excluded items that were designed to identify sources of involvement (i.e. antecedents of involvement), those items that had been identified as having presented confounding issues and the attitude items from the original scale. Mittal (1995) argued that this reduced the scale to items that operationalised involvement rather than identifying sources of involvement.

Mittal's (1995) revision of the PII was chosen because it was designed to measure overall involvement, in this case in biosecurity. This scale is short, containing only five items. This helped address several issues, highlighted by McQuarrie & Munson (1987), including reducing respondent fatigue and reducing the length of a survey with an involvement scale. The five-item scale is outlined in Table 1.

Table 1  Mittal's (1995) five-item scale for measuring involvement (items marked with an asterisk needed to be reversed scored).

Important to me	1	2	3	4	5	Unimportant to me*
Of no concern to me	1	2	3	4	5	Of concern to me
Means a lot to me	1	2	3	4	5	Means nothing to me*
Matters to me	1	2	3	4	5	Does not matter to me*
Significant	1	2	3	4	5	Insignificant*

In order to cover involvement in the range of issues covered by the term biosecurity, five aspects of biosecurity were identified. Each was then defined and checked with scientific experts. The five descriptions were biosecurity, quarantine, invasive animal species, invasive insects, and exotic diseases of plants, animals and humans. In each case, the definition for each area of biosecurity was provided as is shown in Table 2, followed by a question. For each question, respondents were required to indicate their level of involvement with that aspect of biosecurity using the scale outlined in Table 1.

Table 2  Definitions questions for involvement in five aspects of biosecurity.

Aspect of biosecurity	Definition and Question
Biosecurity	Biosecurity is defined as keeping out, getting rid of or managing risks posed by pests or diseases to the economy, environment and human health. Biosecurity is:
Quarantine	Quarantine is defined as the process of trying to minimise risk of exotic pests and diseases entering a region. Quarantine issues are:
Invasive animal species	Invasive animal species are defined as animals native to another region that spread widely and cause harm in another region. Invasive species are:
Invasive insects	Invasive insects are defined as insects native to another region that spread widely and cause harm in another region. Invasive insects are:
Exotic diseases of plants, animals and humans	Exotic diseases of plants, animals and humans are defined as any disease or strain of a disease that is new to New Zealand. Often these diseases have the potential to spread quickly and cause severe problems and/or death to the plant, animal or person that catches it. Exotic diseases of plants, animals and humans are:

A respondent with the lowest level of involvement in a particular aspect would have a score of 5 and a respondent with the highest level of involvement would have a score of 25. The overall measurement of involvement is the average score across all aspects of biosecurity. As such, the highest level of involvement is an overall score of 25.

Perceptions of biosecurity

In addition, participants were asked to indicate their overall view of biosecurity by responding to several statements about biosecurity. The statements were adapted from the work of Obermiller (1995) who used these statements to determine perceived control, concern and importance of water and energy conservation and recycling and solid waste reduction. The statements, outlined in Table 3, were rated on a five-point scale, where 1=strongly disagree and 5=strongly agree. The responses to these scales provided a comparison with the calculated level of involvement from Mittal's scale.

Table 3  Statements used to indicate participants' overall view of biosecurity (five-point scale; 1=strongly disagree, 5=strongly agree).

1	There is not much one individual can do about biosecurity
2	The effort of one person to declare risk goods is useless as long as other people refuse to declare risk goods
3	The biosecurity risk to New Zealand is exaggerated, in the long run things balance out
4	I don't think New Zealand biosecurity is very important
5	The potential seriousness of biosecurity is frightening

Recall and response to biosecurity information

Respondents who had travelled back into New Zealand within the 12 months prior to the study were asked a series of questions about their response to a range of biosecurity requirements and information. As level of involvement was thought to influence the amount of information search and effort put into decision making, some indication of the biosecurity information individuals had seen and noticed, and their response to that information, was required. Respondents were asked to indicate if they had seen information in the form of:

•	a brochure/pamphlet on biosecurity
•	a video shown on board most inbound international aircraft
•	the New Zealand passenger arrival card
•	amnesty bins
•	other information and signs in the international arrivals hall.

Respondents were also asked if they had seen detector dogs in the international arrivals hall. Respondents were also asked to rate the effect of the information provided using three criteria, whether the information changed their mind about declaring risk goods, the usefulness of the information and the importance of the information in highlighting New Zealand biosecurity requirements.

A web-based survey was used to collect data. SurveyPro 4, developed by Apian Software, was used to design the survey. The initial set of questions was put together in a Word file. This was sent to an IT consultant who used it as a template to set up the survey in SurveyPro. The survey was published and piloted by six colleagues. Some small changes were highlighted required and these were incorporated into the survey.

Target audience

The focus of the research was to understand individuals’ response to New Zealand biosecurity requirements. A convenience sample of postgraduates and staff at Lincoln University, as well as researchers at two research institutions and participants at South Island field days hosted by Lincoln University, were used to explore involvement with biosecurity. This decision was based on the characteristics required of respondents, particularly in terms of accessibility, and because the population at the university and research institutions is generally diverse, comprising young adults, adults, international students and staff born and trained overseas. This convenience sample included individuals from a range of countries of birth, individuals who had travelled overseas within the last 12 months and individuals with a basic understanding of biosecurity; there was thus the potential for a range of levels of involvement among these individuals.

Other researchers have used university students as convenience samples to investigate involvement, notably Gregory (2004), Mittal (1995), Roser (1990) and Zaichkowsky (1985), and all have found these populations useful. In the literature on social research in biosecurity, researchers have focused on particular groups of people, such as visitors to sites similar to the field days. Hall (2003) surveyed visitors to vineyards in three locations and García-Llorente et al. (2008) surveyed residents and visitors within a particular region.

The survey was distributed via email to postgraduate students from Lincoln University in August/September 2008 (sample 1). Emails were sent via faculty staff inviting postgraduates to complete the survey. No incentive to complete the survey was provided. A reminder email was sent approximately 2 weeks later. The response rate was low, with only 49 responses from a population of approximately 600, of which 43 were usable.

A second round of surveying was undertaken between February and April 2009. For this round of surveying, the target audience consisted of staff at Lincoln University and two research institutions within New Zealand, as well as participants at the South Island field days at Lincoln University (sample 2). Staff at Lincoln University and staff in sections with the research institutions were emailed, inviting them to complete the survey. No incentive to complete the survey was provided. At the field days, attendees who visited the Lincoln University site were asked to complete the survey. Again the response rate was low, with 85 responses, of which 82 were usable. This provided a total of 125 usable responses. Individual respondents were not able to be identified; however, the date when the survey was completed was recorded so participants from the two rounds of surveying could be separated during analysis. Data were analysed using the Minitab software package; χ ², Mann–Wallis and Friedman tests were used to determine significant differences in the data as appropriate.

Results

As already noted, a total of 134 responses were received over two rounds of web surveying, providing a total of 125 usable responses. The respondents were individuals who worked or were associated with agriculture and land based industries. Approximately 37% of respondents indicated that they dealt with biosecurity issues at work. The majority of respondents were under 50 years of age, roughly evenly spread between male and female (52% male, 46% female), and slightly over half were born in New Zealand or Australia (56%). Respondents from the first round of surveying were significantly younger than those in the second round, with 51% indicating they were 18–30 years old and another 26% indicating they were 31–40 years old. This is not altogether surprising given that the target audience in the first round of surveying were postgraduates.

Measuring involvement in biosecurity

Overall involvement in biosecurity was calculated using Mittal's (1995) adaption of the PII. Cronbach's alpha was used to determine the internal reliability of the scale for each item. Mittal (1995) reported an internal reliability of 0.90. In this case, the Cronbach's alpha for each aspect of biosecurity was 0.80, 0.87, 0.86, 0.86 and 0.86 respectively, indicating an acceptable level of internal reliability. Four of the five items on each scale were reverse scored. These were re-coded and a total level of involvement calculated by adding the five items together. This provided an involvement score between 5 (the lowest level of involvement) and 25 (the highest level of involvement) for each of the five areas of biosecurity defined in the survey. The average involvement score across these five areas was calculated and individuals were grouped into one of four categories, based on their score. The categories were determined in a manner similar to that used by Zaichkowsky (1985), by exploring the distribution of involvement scores, and assessing the ratings needed to obtain those scores. Low involvement was categorised as a score from 5 to 11; medium involvement, a score from 12 to 18; high involvement, a score from 19 to 24; and very high involvement was categorised as a score of 25. As can be seen in Table 4, over half of respondents had high involvement, with another 23% indicating they were very highly involved in biosecurity.

Table 4  Level of involvement of respondents based on average involvement scores across five areas of biosecurity.

Level of involvement	Total number of respondents	%
Low (5–11)	2	2
Medium (12–18)	26	21
High (19–24)	68	54
Very high (25)	29	23
Total	125	100

The data were analysed to determine whether there were any significant differences in demographics of these groups. The two individuals who had low involvement were excluded from this and any further analysis because it was not possible to undertake any valid statistical comparison with only two respondents. The only significant difference between respondents with different levels of involvement was their age (χ ²=14.737, P=0.022), outlined in Table 5. 45% of respondents with very high levels of involvement were under 30 years of age, compared with 25% highly involved and 31% with medium involvement. 37% of respondents with high levels of involvement were over 50, compared with 17% who were very highly involved and 12% with medium involvement. There were no differences in gender, whether biosecurity issues were dealt with at work or the number of times travelled. This result is similar to that predicted by Diamantopoulos et al. (2003) who argued that demographics alone are a poor means of defining an individual with regard to an issue such as biosecurity.

Table 5  Age distribution for each level of involvement*.

	Level of involvement
Age (years)	Medium	High	Very high	Total
18–30	8	17	13	39
31–40	11	11	7	29
41–50	4	15	4	24
51+	3	25	5	33
Total (n)	26	68	29	125
*Significantly different; χ ^2 =14.737, P=0.022.

Extent of information processing

The data were also analysed to determine whether there were any differences in the response to the statements about biosecurity based on level of involvement. Generally, those individuals with very high involvement had significantly different perceptions from those individuals with medium involvement (see Table 6). Individuals who were very highly involved were more likely to agree with the statement that the potential seriousness of biosecurity was frightening (mean score 4, where 1=strongly disagree and 5=strongly agree) and disagree with statements that biosecurity was not important (mean score 1), not much one person could do (mean score 1.1), that the risk is exaggerated (mean score 1.3) and that the effort of one person to declare risk goods is useless (mean score 1.4). These results confirmed that these individuals were very highly involved with the issue of biosecurity.

Table 6  Mean score for statements on biosecurity depending on level of involvement (low, medium or high) (1=strongly disagree, 5=strongly agree).

	Involvement
	Medium	High	Very high
I don't think New Zealand biosecurity is very important^*	1.5	1.5	1.0
There is not much one individual can do about biosecurity	1.8^a	1.4	1.1^a
The biosecurity risk to New Zealand is exaggerated, in the long run things balance out	2.3^bc	1.6^b	1.3^c
The effort of one person to declare risk goods is useless as long as other people refuse to declare risk goods	2.1^d	2.3^e	1.4^de
The potential seriousness of biosecurity is frightening	3.6^f	3.9	4.0^f
*Mann–Whitney tests were unable to undertaken on this statement as all individuals in the highly involved category had rated this statement 1 (meaning that all values in the column were identical). Significant difference: ^a P=0.0022; ^b P=0.0026; ^c P=0.0002; ^d P=0.0026; ^e P=0.0007; ^f P=0.0278.

Respondents were asked about a range of information provided on biosecurity when travelling into New Zealand. There were no significant differences between those who had medium involvement, those with high involvement and those with very high involvement in terms of the information they had seen.

There were some significant differences in the rating of different types of biosecurity information when comparing medium, high and very high involvement respondents. Those who were very highly involved in biosecurity were more likely to indicate that the arrival card was useful than those individuals who had high or medium involvement. Otherwise, there were no significant differences in the ratings of usefulness.

There were some significant differences in ratings of whether information provided changed an individual's mind about declaring risk goods. Those respondents with a medium level of involvement were more likely to be neutral about whether any of the information provided changed their mind about declaring risk goods. Individuals who were less involved were more likely to indicate that the information changed their mind about declaring risk goods.

There was no difference between medium, high and very high involvement respondents in their rating of the importance of biosecurity information. Overall, those respondents who were very highly involved rated information as being more useful, but not in terms of changing their mind. Individuals who were less involved were more likely to indicate that the information changed their mind in terms of declaring risk goods, but were less inclined to rate the information as useful.

The results presented to this point indicate that it is possible to measure involvement in biosecurity using Mittal's (1995) revised PII. High involvement in biosecurity was confirmed through the response to statements designed to infer respondents’ overall view of biosecurity. Respondents indicated that they felt biosecurity was important and they were concerned about it. This is consistent with high involvement. However, those who were highly involved in biosecurity did not appear to take more notice of information (compared with those who were less involved), although there were some differences between respondents in terms of their rating of biosecurity information. These results are in contrast to the literature on involvement, which indicates that those who are more involved tend to pay more attention to information and put more effort into processing that information. This could be a critical point of difference between being involved in an issue such as biosecurity compared with involvement in terms of consumer behaviour (purchasing a product).

Discussion

The results of this study indicate that respondents were highly involved in biosecurity. The scale used to measure involvement in this issue was a scale originally developed for consumer products. However Mittal's (1995) involvement scale, a revision of the PII developed by Zaichkowsky (1985), appeared to provide a robust measure of the level of involvement in biosecurity. This level of involvement was validated via participants' responses to a range of statements about biosecurity indicating they were concerned about biosecurity and felt it was important. Respondents’ backgrounds also indicated they should be involved with biosecurity, working in industries that had potential links to biosecurity issues and dealing with biosecurity issues at work. Younger people appeared to be more involved in biosecurity. It is not entirely clear why this was the case.

However, the results of this study indicate that there were differences between involvement in this issue and involvement in consumer products. Much of the literature on involvement centres on consumer products and advertising (Assael et al. 2007; Zaichkowsky 1994). The literature on involvement suggests that those who are highly involved should allocate time and effort to searching for information in order to make a decision about a product or issue, for example by reading widely or seeking information from experts (Celsi & Olson 1988; Flynn & Goldsmith 1993; Lee et al. 1999; Zaichkowsky 1994). Respondents who were highly involved did not indicate that they read or took note of more information than those who were less involved. However, very highly involved respondents rated some biosecurity information differently to those who were less involved. Very highly involved respondents indicated that they had already made up their mind and were less likely to indicate that information changed their decision to declare risk goods.

There is some evidence in the literature to suggest that, once attitudes are formed, individuals who are highly involved are less likely to process information that is counter to those attitudes (Park et al. 2007). However, while attitudes are being formed, highly involved individuals are more likely to take note of information relevant to the issue or product (Park et al. 2007; Priluck & Till 2004). The implication for the results of this study is that very highly involved respondents had already formed beliefs and attitudes about New Zealand biosecurity requirements and so were not inclined to take notice of the information provided to them as they came back into the country. This suggests that these individuals already felt they knew what they needed to do to comply with biosecurity requirements.

Limited information search under high-involvement conditions has been explored by Moorthy et al. (1997). They presented evidence to suggest why the relationship between the amount of search and experience with the product can be an inverted U shape. In their analysis, the consumer becomes more like an expert as experience increases, decreasing the cost of searching for information, while increasing the opportunity cost of time (Moorthy et al. 1997). Travellers in our sample had apparently decided that they had gathered enough information to be able to meet the biosecurity requirements and so the opportunity cost of further information search was high, given they felt the environment was stable and the requirements should not change dramatically.

This type of behaviour, limited search for information under high-involvement conditions, is exemplified in the consumer decision making process of brand loyalty. Brand loyalty is characterised by high involvement but with less effort put into information search (Assael et al. 2007). As consumers become more experienced with a product or product class, loyalty tends to increase and information search decreases (Ratchford 2001). Brand loyalty tends to occur when a purchase decision is considered risky, or is a source of self-identification, such as when buying a vehicle or a house (Assael et al. 2007; Richins et al. 1992). Assael et al. (2007, p. 127) describe brand loyalty as instrumental conditioning, ‘positive reinforcement based on satisfaction with the brand, leading to repetitive behaviour’. The results of the research outlined in this paper appear to suggest that respondents were exhibiting a type of brand loyal behaviour with regard to biosecurity, stemming from enduring involvement with the issue.

Enduring involvement has been explored to some extent in the literature, including the impact on the search for information over time. Research on involvement in leisure activities would suggest that individuals can have enduring involvement in recreational behaviour such as camping (McIntyre 1989). Research on involvement in food has revealed that, although traditionally assumed to be a low-involvement product, some individuals exhibit high and enduring involvement in this product category (Verbeke & Vackier 2004). Verbeke & Vackier (2004) found that consumers of meat had varying levels of involvement in that product. Although all were concerned about tangible qualities such as taste, consumers with high involvement were also concerned about intangible characteristics of the product such as quality in relation to health and food safety. Some consumers have enduring involvement in food, perhaps because of their concerns about food safety, as mentioned above, or because of a perceived need to support sustainable food production (Vermeir & Verbeke 2006). While involvement in food continues to be high over time, this may not necessarily translate into an extensive search for information and processing. For example, once a product is found that meets the criteria required, the consumer simply continues to purchase this product and there is little need for any extra information.

To some extent, with regard to biosecurity, this type of behaviour is desirable. Travellers who have put time and effort into finding out about the biosecurity requirements for New Zealand and then follow those requirements will present the least risk in terms of biosecurity. However, there is a danger that those individuals who are highly involved may have an unfavourable reaction to the biosecurity requirements presented. Under these circumstances, there is the possibility that they then may actively work to try and avoid the biosecurity requirements. This type of behaviour has been highlighted by Kaine et al. (2010) in their exploration of people's response to regulations.

Another undesirable aspect to brand loyal behaviour is that individuals may not become aware of changes to biosecurity requirements since they do not perceive any need to gather further information and, as a result, may not take notice of any changes. This could be a problem, depending on the change in the requirements. Being able to signal to highly involved travellers that they need to pay attention to some new information could be an important part of ensuring that they remain a low-risk group.

Limitations of this research

One of the key limitations of this research was that the sample was small and consisted of respondents involved in agriculture and land based industries. Most respondents had a high level of involvement in biosecurity. The results can thus only provide a detailed understanding of high involvement in biosecurity among individuals from this sector. The survey was not designed to obtain a representative sample from agriculture and land based industries. However, there is no reason to suspect that the information is not representative of individuals in this sector.

Obtaining responses from individuals with low involvement is difficult. By definition, individuals with low involvement are less engaged and less inclined to seek out information (Illies & Reiter-Palmon 2004) and so may not be inclined to respond to a survey. The authors of one of the few studies on involvement and response to surveys concluded that high involvement in a topic led to a higher response rate, however the speed of response did not differ (Van Kenhove et al. 2002). Others have speculated that willingness to respond to a survey could be an indicator of high involvement (Kinard & Capella 2006). This suggests that providing incentives or rewards would be a means of obtaining responses from individuals with low involvement in the issue. No incentives or rewards were offered to potential participants in this research.

A second limitation of this research is that respondents were asked to recall information on biosecurity that they had seen when coming back into New Zealand sometime in the previous 12 months. Self-reported recall of information can be problematic, in terms of both under- and over-reporting of information recalled. If individuals believe that their answers could reflect a negative image, then over- or under-reporting can be an issue. Kopcha & Sullivan (2007) found that teachers tended to report greater personal use of a range of teaching practices in the classroom than actually occurred. Research into food intake tends to have an under-reporting bias as individuals either do not wish to be noted as consuming more food than average or because they alter their diet during the period in which they are surveyed (Cook et al. 2000; Pryer et al. 1997). In addition, memories about oneself can tend to be inflated or elevated, exaggerating current performance (Gramzow & Willard 2006; Willard & Gramzow 2008). Some or all of the issues outlined could have influenced the response of individuals, and thus the results of the research outlined in this paper.

A third limitation to this study is that the source of involvement in biosecurity was not able to be explored because involvement was treated as a single dimension construct and measured using a quantitative scale. Mittal's (1995) revision of the PII was used to produce a one-dimensional measure of involvement that was easy to administer in a survey. However, other researchers (e.g. Michaelidou & Dibb 2006) have highlighted the need to consider the different dimensions of involvement identified by Kapferer & Laurent (1985). Some researchers have successfully used a qualitative approach to exploring involvement and sources of involvement or have used longer quantitative involvement scales. Aldoory & Van Dyke's (2006) study used qualitative methods to explore involvement and response to a bioterrorism attack on food supplies. They found that highly involved participants were more inclined to talk about searching for information and were able to identify factors that increased involvement in this issue. Taking a quantitative approach, Michaelidou & Dibb (2006) were able to identify different sources of involvement in clothing from responses to a 15-item involvement scale that was part of a web-based survey.

Implications

There are several implications of the results of this research into individuals and their response to New Zealand biosecurity requirements. The first implication is that highly involved individuals had an understanding of New Zealand biosecurity requirements and the implications of not meeting those requirements. In many respects, this demonstrates that the time and effort that has been put into the design and delivery of information to travellers has been working (Rauniyar et al. 1999, 2000; Whyte 2005), at least for those individuals who feel that biosecurity is important.

The nature of the sample of respondents in this study meant that the response of individuals with low involvement could not be elicited. Marketers emphasise the need to have a different strategy for gaining the attention of individuals with low involvement (Assael et al. 2007). Assael et al. (2007) suggest that repetition, focusing on a few key points, and emphasising visual components are keys to attracting the attention of low-involvement consumers. In addition, Kim (2003) suggested that an affect-evoking strategy, particularly the use of fear, was the most effective means of attracting the attention of individuals with low involvement. Kim (2003) also suggested an alternative strategy, cue-emphasising, when enduring involvement was low but situational involvement was high. An individual's response to biosecurity could be in this category, as having to answer questions about biosecurity and have baggage X-rayed could evoke situational involvement in the issue. In these circumstances, Kim (2003) suggests using a celebrity to emphasise the need to declare biosecurity risk goods.

Interestingly, until recently, this had been the approach taken by the Australian biosecurity authorities. Steve Irwin was the public face of the ‘Quarantine Matters’ communication campaign from 2002 (Smitz et al. 2004; Thompson et al. 2009). In this campaign Steve Irwin discussed the need to declare biosecurity risk goods in order to protect Australian flora and fauna, and emphasised that authorities would find risk goods if individuals did not declare them. This combined both the affect-evoking and cue-emphasising strategies outlined by Kim (2003). Steve Irwin's death prevented this campaign from continuing, although there has been some disagreement about its effectiveness (Thompson et al. 2009).

An alternative, longer term, strategy is to try and increase the level of involvement in biosecurity. In work exploring involvement in health care, Roth (1994) emphasised the need to invoke trust, control and empowerment in individuals as a means of increasing their involvement. In the health care situation this meant increasing the means by which individuals could have contact with, and obtain information from, health care professionals. Assael et al. (2007) also recommended a number of methods for increasing involvement in the context of consumer goods, including considering linking the issue to another more involving issue or situation. For biosecurity this could mean focusing more on the economic or environmental impacts of biosecurity, depending on which is more important to an individual. The participants in this research tended to be involved in agriculture or be associated with primary industries. Biosecurity incursions can have a devastating impact on these industries, so linking the importance of declaring risk goods at the border to protecting an industry from the impact of a biosecurity incursion could increase involvement. For others, the impact of a biosecurity incursion on their experience of the natural environment could help increase their level of involvement in this issue.

The second implication arising from this research was that those individuals highly involved in biosecurity could reach a point at which they felt they did not need to process more information on this issue. This is not necessarily a problem. It means that most of these individuals should understand biosecurity requirements and follow them. However, it is an issue if biosecurity requirements change. Ensuring that these high-involvement individuals take note of new information could prove difficult, especially if they believe they already know what is required. Providing information to people when they have limited their information search to some extent resembles low-involvement behaviour. In these circumstances, the strategies outlined above will also be relevant if there is a need to try and regain individuals’ attention.

Finally, this research demonstrates that understanding involvement in an issue can provide significant information on an individual's behaviour in regard to that issue. When outlining initial forays into research into involvement, Krugman (1965, p. 355) stated ‘The significance of conditions of low or high involvement is not that one is better than the other, but that the processes of communication impact are different’. This succinctly sums up the significance of involvement; that is, it is important to know so that you can target information to your audience effectively.

Future research

There are a number of opportunities for further research based on the results outlined here. The study could be extended by surveying a larger sample of the New Zealand population to determine whether there are any differences in the results and thus the requirements for successful design and communication of biosecurity information. The study could also be extended to specifically explore other groups, for example, foreign tourists, or subsets of tourists such as trampers or bus tour participants, in order to explore their level of involvement in biosecurity.

Conclusion

The New Zealand biosecurity system has been designed to exclude, eradicate or manage the risks posed by pests or diseases to the economy, environment and human health (Biosecurity Council 2003). People are a fundamental part of the system, representing both an opportunity and challenge. The research outlined in this paper was designed to explore a small part of the biosecurity system: individuals’ responses to New Zealand biosecurity requirements. The results indicate that highly involved individuals have taken note of, and appear to respond to, biosecurity requirements at the border. Successful communication with this group of people means highlighting when the requirements may have changed and when they need to take note of the information provided. This could take the form of highlighting the consequences of not adhering to the biosecurity requirements and providing details of any changes in requirements. Effectively communicating to individuals with low involvement appears to be more of a challenge. Attempting to increase their level of involvement could be a potential strategy, but more research is needed to determine the most appropriate means of doing this.

Acknowledgements

Thanks go to the postgraduates and researchers who participated in this survey. The authors are also grateful to the Lincoln University Marketing, Communications and Student Recruitment group who kindly allowed us space in the Lincoln University tent at the 2009 South Island field days to recruit more respondents for the survey. This work was partly funded by the New Zealand Foundation for Research, Science & Technology through contract CO2X0501, the Better Border Biosecurity (B3) programme (www.b3nz.org).