Understanding Insurance Knowledge: A Brief 7-Item Measure

The conceptualisation of the insurance culture and the identification of a standard measuring instrument are the first steps towards defining a literate consumer. Obviously knowledge, understood as the ability to understand and use concepts in a conscious way, is considered to be a key variable for measuring the levels of many conceptual definitions of literacy, so also for insurance literacy. The aim of our research is precisely to verify the validity and reliability of a questionnaire that is composed of 7 questions that can represent a tool for measuring the level of insurance knowledge of consumers. The questions investigate the mere knowledge of insurance definitions and concepts, without going into too much detail about specific types of policies. To pursue this goal, a factor analysis has been conducted through a sample that is composed of 274 Italian respondents. The results show that those items are able to measure the basic insurance knowledge of a consumer.


Introduction
Financial literacy, defined as "knowledge and understanding of financial concepts and risks, and the skills, motivation and confidence to apply such knowledge and understanding in order to make effective decisions across a range of financial contexts, to improve the financial well-being of individuals and society, and to enable participation in economic life" (OECD 2014), is the key to making informed financial decisions. Similarly, insurance literacy enables consumers to make insurance decisions.
While financial literacy has been studied and deepened from many points of view, studies on insurance literacy are still limited. One possible justification for not developing this line of research is the lack of official criteria to classify a "literate" consumer (Tennyson, 2011).
To measure a consumer's level of insurance literacy, as well as financial literacy, multiple variables can play a key role. Among these, insurance knowledge, understood as the ability to understand and use insurance concepts in a conscious way, is the discriminating factor that helps the consumer to make conscious and specific choices for his or her needs. According to some researchers (Gine, Townsend & Vickery, 2008;Tennyson, 2011;Cole, Gine, Tobacman, Topalova, Townsend & Vickery, 2013;Driver, Brimble, Freudenberg & Hunt, 2018;Lin, Bruhn & William, 2019), consumers without insurance knowledge and the important role it plays do not consider the insurance policy a risk management tool. For this reason, consumers often tend not to insure themselves, or to insure themselves inadequately, and are thus under-insured when an adverse event occurs (Rice, 2016;Lin et al., 2019Lin et al., ). 2014ANZ, 2015), institutions around the world have attempted to remedy the problem by imposing rules and regulations for intermediaries to follow. For example, the Insurance Distribution Directive, the new European Directive on the distribution of life and non-life insurance products, has recently been introduced in Europe. This is a new set of rules aimed at placing the actual needs of the client at the centre of the logic of product construction and distribution. This operating mode is possible thanks to the revision of pre-contractual and contractual documentation in order to allow the consumer comparability between products, profiling of actual needs, and greater transparency.
Some research has shown that insurance policy information materials are written in a complex and difficult to understand way (Pati et al., 2012;Vardell, 2013). In addition to guaranteeing transparency, some American states have imposed "simple language" requirements aimed at reducing the use of courtly and complex language in favour of commonly used terms (Stempel, 2006). In addition to this, some states have imposed a minimum requirement for the readability of insurance contracts. (NAIC, 2010). The readability of a contract is measured by an index called the Flesch Reading Ease Score, which is determined by applying an algebraic formula using syllable count and sentence length in the contract. Despite the efforts made by the scientific community and institutions to address this problem to date, not only does the problem persist, but it has not yet been analysed and resolved at the source; i.e., it is still unclear what discriminators determine a consumer who is not literate in the insurance field and consequently what impact this has on his or her decision whether or not to take out insurance against certain unfavourable events. It is necessary to create a tool for measuring insurance literacy.
Our contribution to scientific research consists in the creation of a tool for measuring insurance knowledge. This instrument actually consists of a questionnaire of 7 multiple-choice questions, that is useful for determining the level of insurance knowledge of an individual and also able to highlight what are the conceptual gaps of respondents. The questions investigate the mere knowledge of insurance definitions and concepts, without going into too much detail about specific types of policies; for each question there is only one correct answer. To test the usability and thus the replicability of the tool, we conducted an exploratory factorial analysis of 274 observations. The work is organised as follows. In section 2, we present the literature review, in section 3, we present the methodology, in section 4, we present the results, and in section 5, we outline our conclusions.
The conceptualisation of the insurance culture and the identification of a standard measuring instrument are the first step towards defining a literate consumer.
The aim of our research is precisely to create, through the use of a questionnaire, a tool for measuring insurance basic knowledge of consumers without going into too much detail about specific types of policies. To do this, first of all, it was decided to use the methodology of the literature review as it can highlight any gaps in the literature (Baggio & Valeri, 2020). For this reason, the work previously carried out by other researchers has been analysed.
Through the main electronic databases, the most relevant studies have been selected through questionnaires bearing the following keywords: "Insurance Literacy", "Financial and Insurance Literacy", "Financial and Insurance Education", "Insurance Knowledge".
As shown in Table 1, the studies that dealt with the issue of insurance literacy through a questionnaire amounted to 21. Notes: *knowledge also in terms of insurance function; **rating also refers to studies in which the answers were analysed in percentage terms The table shows that most of the previous studies (85.7%) focused on insurance questions specific to a particular non-life or life-line of business, while only 4 out of 21 studies asked questions about general insurance concepts that could not be subject to a specific instrument.
A number of studies have asked questions about finance, as previous studies on financial literacy have identified insurance as a particular category of finance itself (Huston, 2010;Remund, 2010;Zait & Bertea, 2014;Lin et al., 2019), although Lin et al. (2019) show that financial literacy does not necessarily translate into insurance literacy.
Overall, there are two shortcomings which, in our view, have affected the results of previous research. The first, which brings all studies together and at the same time differentiates them in the research objective, is the lack of a definition of insurance literacy to refer to.
In general, the concept of literacy refers to the notions, knowledge and skills needed to understand and use information related to a specific topic.
According to Sampath Sanjeewa and Hongbing (2019), insurance literacy can be defined as a combination of knowledge, skills, attitudes, and behaviour necessary to make an informed insurance decision based on potential risk exposure and individual circumstances.
Obviously, knowledge is the most common component of many conceptual definitions of literacy.

Understanding potential risk exposure, 2). Risk mitigation strategies, 3). Insurance concept, principles and benefit, 4). Insurance products and covers 5). Rights and duties of insured, and 6). Information sources".
We believe that insurance literacy is a construct composed of 3 components: insurance knowledge, insurance understanding and insurance skills and attitudes (Allodi, Cervellati & Stella, 2020).
These three components determine a consumer's degree of literacy and help him or her to make part of his or her decision about whether or not to buy an insurance product or service. However, this is not sufficient because the consumer's final decision depends on an intrinsic behavioural component of each individual that cannot be included in the concept of insurance literacy. We believe that knowledge is the component that plays a ijbm.ccsenet.org International Journal of Business and Management Vol. 16, No. 2;2021 predominant role in determining the level of insurance literacy of a consumer, so we think that knowledge should be measured by distinguishing between the simplest knowledge, regarding insurance concepts and principles in general, and advanced knowledge, regarding the characteristics of the various products and lines of business.
This differentiation of knowledge is the second shortcoming of the studies carried out so far, i.e., the lack of a measurement indicator that considers an assessment of the level of insurance knowledge by distinguishing between basic and advanced level questions. Only Tennyson (2011) proposes a questionnaire with questions referring both to specific products and to general concepts of the insurance world, but he measures insurance literacy through the percentage analysis of the answers to the questions, referring indiscriminately to the concept of insurance literacy and insurance knowledge.
In order to solve this problem, we propose a new questionnaire that takes into account the shortcomings exposed and focuses its questions on basic insurance knowledge.

Description of the Survey
The focus of our research refers to the relation between knowledge and insurance among Italian citizens. To do so, we designed a questionnaire. The assessment was used to measure the levels of Insurance knowledge. The survey was fielded in November 2019. The data were collected during the period January-February 2020 through an online survey. The sample is composed of 274 Italian respondents. We can define our sample as a convenience sample, as several university students and former students, relatives and friends were involved, and a voluntary response sample cause the link to access to the questionnaire was posted online through the groups of the main social network.
The survey reports information on several demographic characteristics such as age, gender, and education. In Table 2, we describe the variables that we use and how we codified them for the analyses that follow. Education 0-9 Years =1; 9-12 Years = 2; More than 12 Years =3 Work status Unemployed = 0; Student = 1; Retired = 2; Part-time work (< 35 hours/week) = 3; Full-time work (> 35 hours/week) =4

Measures
In this work, seven questions have been tested to measure insurance knowledge. In particular, these questions were developed to assess financial insurance concepts. The correct option is scored as 1 and the wrong option as zero. Finally, following the methodology in this research, the response "I do not know" and "Prefer not to say" are considered wrong. (Table 3).  Table 4 illustrates the descriptive statistics for the sample of respondents to the questionnaire. The age of the respondents ranges from 20 to 90 years. The average age of the respondents is 34.91 years, and the standard deviation (SD) is 11.20.  In Table 5, we present the frequency tables related to the variables available in our dataset. Among socioeconomic characteristics, we considered age, gender, civil status, education levels, work status, and income.

Factor Analysis
We conducted a factor analysis (using SPSS 26 software). As reported in Table 6, the Kaiser-Meyer-Olkin (KMO) index is equal to .789, whereas the Bartlett test is significant (p = .000). These results reject the null hypothesis that the matrix of correlations between variables is an identity matrix (Hair, Anderson, Tatham & Black, 1995).  According to the K1 criterion (Kaiser, 1960), factor analysis proposes the extraction of one factor, which explains in cumulative terms 31.19% of the variance. The number of factors to be extracted for this analysis, equal to one, is confirmed by the scree plot. (Figure 1). Table 7 presents the results from the factor analysis. We considered one factor related to respondents' insurance knowledge. The factor decomposition is implemented via maximum likelihood (ML) factor analysis with the Promax method (Costello & Osborne, 2005). The Pattern matrix shows that there is only one factor that represents the insurance knowledge. Furthermore, Cronbach's alpha for all questions is higher than 0.65. In particular, this result demonstrates that these questions can be used to analyse insurance knowledge.

Conclusions
We propose and validate a new insurance knowledge questionnaire to test one of the three dimensions identified to define the concept of consumer insurance literacy. The econometric analyses conducted on the questionnaire have established how the questions developed on insurance knowledge are able to measure basic insurance knowledge without going into the details of particular types of policies. The reliability analyses carried out on ijbm.ccsenet.org International Journal of Business and Management Vol. 16, No. 2;2021 the questions in the questionnaire that we developed showed that they are reliable and replicable, thanks to the calculation of Cronbach's alpha which is greater than 0.6. The factor analysis demonstrated the robustness of the questionnaire. From the research point of view, this study has some advantages.
First of all, it is a small step forward compared to previous studies. Having a validated questionnaire that allows one to investigate the phenomenon of insurance literacy through the basic knowledge of insurance issues on the part of individuals is an incentive to face more in-depth analysis and to investigate if and how the variables identified as significant for financial literacy also play a fundamental role in the insurance sphere. The replicability of the questionnaire will allow it to be tested on individuals of different generations and ages and above all from culturally different backgrounds; diversity will be a crucial aspect in the evolution of the surveys.
Secondly, one aspect that should not be underestimated is the length of the questionnaire. Since there are only seven insurance knowledge questions, the questionnaire does not take long to complete, and it can be completed and replicated in just five minutes.
However, the research has one limitation-it does not provide a scale for measuring insurance literacy, but only insurance basic knowledge.
Our goal for the future is to expand the questionnaire by including three more sections, i.e. questions on advanced knowledge, understanding, skills and attitude, in order to provide a comprehensive scale for measuring insurance literacy.