When legal rights are not a reality: do individuals know their rights and how can we tell?

Public knowledge of rights has been the subject of a number of empirical enquiries over the last decade. In England and Wales, knowledge of rights and its relationship with an individual's capacity to ‘self-help’ and ‘self-represent’ when faced with a civil justice problem has become the subject of renewed attention following changes to legal aid which, from March 2013, will see the availability of legal advice and representation dramatically reduced. Previous studies focusing on public knowledge of rights in this (and other) jurisdictions have illustrated a lack of knowledge amongst the general population and more specifically, a widespread tendency of individuals to assume that the law aligns with their own moral, ethical or social attitudes. However, many of these studies have also suffered from methodological shortcomings. In attempting to address some of these shortcomings this study uses an open-ended format to ask individuals with one or one or more civil or social justice problems to describe their rights/legal position. We find that whilst an open-ended question approach to exploring knowledge of rights yields insight not acquired by other formats, its utility is constrained by difficulty reconciling articulation and actual knowledge of rights. We discuss the implications of these findings as they relate to the development of future research in the field of family and social welfare law, Public Legal Education (PLE) and access to justice post-March 2013.


Introduction
In 2010 the BritishG overnment proposed wide-ranging reforms to the legal aid system in England and Wales (Ministry of Justice 2011). They notedt hat although 'successive changes( had) managed to contain the growth in overall spending on legal aid, such changes( had) not addressedt he underlying problemsf acing the scheme' (Ministry of Justice2 010b, p. 3). In the long term, the Government agreed that simplification of the justice system to makei ts navigation easier for lay individualsw as necessary; in the meantime, the availability of publiclyf undeds pecialista dvice and representation across ar ange of matter types would be substantially curtailed ( Ministry of Justice 2010a, 2010b. Deciding which areaso fl aw woulda nd would not qualify for publicly fundedl egala dvice and assistance under the new legal aid schemew as justified in part, by reference to whethert he public couldu tilise (self-help and) self-representationf or their particular problem (Ministry of Justice 2010a, 2010b. 1 Consequently, the reforms heraldedb yt he Legal Aid, Sentencing and Punishment of Offenders Act (LASPO) 2012, detailed (explicitly and implicitly) that self-help/self-representation would from April 2013 onwards,b ecome ak ey route to resolving civil justice problems. 2 Public legal capability, education andknowledge of rights The idea that individuals are capable of handling civil justice problems without professional advice, anticipates al evel of existing competence from the populationparticularly those individuals who are neither eligible for legala id, nor able to secure alternative legal services. How public legal capability is assessed and improved, or for that matter, the characteristicsn ecessaryt op roducei ndividual legal 'capability' remains an issue of some debate. One factor generally considered relevant, is the extent to which individuals know their legal rights (see, e.g., PLEAS Taskforce 2007, Gramatikov and Porter 2010, Buck et al.,2008, Balmer et al.,2010, Bowal 1999. It is argued that without such knowledge, individuals will more often fail to vindicate their rights (Bowal 1999); fail to take steps to protect themselvesa gainst the likelihoodo faparticulare ventuality (Kim 1999, Meager et al.,2002; fail to upholdtheir civic responsibilities (Bowal 1999); and experience diminished success whenself-representing in court (Moorhead and Sefton 2005). Al ack of knowledge hasa lso been said to potentially generate unrealistic expectations of lawyers and judges (Bowal 1999) and the risk of experiencing ac ivil justice problem (Williams2009) and is further linked with the adoption of poor problem handling techniques and lower levels of satisfaction with outcome whena ni ndividual seeks to handle their problem alone (Buck et al.,2008, Balmer et al.,2010, Denvir et al., 2012. In the Public Bill Committee debates regarding the passage of the LASPO Bill, it was said that changes to legal aid demanded greater public knowledge of rights if individuals were to be expected to self-help and self-representm ore often (Public Bill Committee 2011). It follows, then, that as part of an on-going effort to safeguard access to justice, whetherthe general public have knowledge of their rights or not will remainone of the key issues facing publicp olicy makers in the years to come. Studies which have previously sought to measure knowledge of rights have found that citizens in an umber of jurisdictions frequently admit to lacking knowledge (see, e.g., Cortese 1966, Williams and Hall 1972, Saunders1 975, Baker and Emery 1993, Parle/IARS 2009, Casebourne et al. 2006Genn et al.,2006, Tennant et al. 2006, Albrecht and Green 1977 with this lackof knowledge more prominent among certains ubsections of the population, namely young people, the mentally ill, and those without educational qualifications (see, e.g., Williams and Hall 1972, Ruck et al.,1998, Youth Access 2002, Parle/IARS, 2009, Casebourne et al. 2006. These concerns have been the basisf or a range of policy initiatives aimed at improving public knowledge of rights, referred to as 'Public Legal Education' (PLE) (PLEAS Taskforce,2 007). 3 Such interventions seek to rectify an absence of knowledge, and thus require an understanding of the existing level of knowledge amongst the general population. Yet measuring public knowledge of rights or for that matter the benefitst hat may flow from increased knowledge is no easy feat, as existing researchint he field has illustrated.

Self-reporting knowledge
Existing studies have generally taken one of two approachesi nm easuring public knowledge of rights.T he first approach, as used by both Denvir et al.,( 2012) and Buck et al.,(2008) has involved asking respondents to self-reportthe extent to which they know their rights,w ith results highlighting low levels of knowledge amongst the general population and certaing roups.H owever, as recognised in the methodological literature (see, e.g., Baldwin2009) whilst self-reporting is aconvenient approach, it is not without issues. These include the fact that long reference periods in any type of survey may impact upon memoryr ecall (Deming 1950, Bradburn et al.,1 987), that respondents to face-toface surveys may be inclined to answer in certainways(for avarietyofreasons, see, e.g., Groves et al.,2009, Calahan 1968, and that the acquisition of knowledge after the event may lead peopletoincorporate post-event information in their memoryofthe experience (Groves et al.,2 009). This lasti ssue has particularr elevance given that an individual's propensity to incorporate post-event information has ramifications for any attempt to link the presence of knowledge with the problem-resolution strategies individuals adopt. More problematic however, is the assumption that thosewho report knowing their 'legal rights' actually do. In the absence of an understanding of whatrespondentsactually believe their rights to be, Buck et al.,( 2008) and Denvir et al.,( 2012) may simplyi ndicate an individual's level of confidence in their knowledge, especially given that existing research also highlights individualsa nswer more confidently whena sked fixed choice questions (see, e.g., Schuman andPresser 1981, Schwarz andHippler 1991).

Fact-basedq uestions with/without ac ontextual hypothetical scenario or 'vignette'
Studies which have attempted to verify the self-reporting of knowledge such as Meager et al.'s (2000) employmentrights research, have adopted asecond approach. Using selfassessment questions alongsideanumber of other question types, including (prompted and unprompted) awareness of various employment related legislation, and questions which ask whether ah ypotheticals ituationr epresents ab reach of the law alongside requiring participants to name the law breached. Via thesemethods, Meager et al.,(2002) were able to demonstrate that although two-thirds of respondents to their study claimed to be well informed or very well informed about their rights, substantivek nowledge varied by individual problem-type, with individuals often able to identify ab reach of the law in a hypothetical situation, but unable to substantiate this breach by naming the relevant legislation. 4 Thea uthors concluded that individuals often identified breaches of the law based on perceptions of fairness or natural justice, rather than knowledge of the relevant legislation (Meager et al.,2002) suggesting that self-reporting remained acrude proxy for actual knowledge.
Yet, whilst utilising arange of approaches, the authors failed to comment on the fact that the ability to namel egislation was not necessarily ag ood measureo fk nowledge of rights or of the content of the legislation itself. By focusing more on whether respondents were able to link what they perceived to be unfair behaviour with the corresponding legislation, rather than exploring whetherrespondents actually understood their rights,the results provided only limited insight into 'knowledge of rights'. Similar such limitations were evident in Kim'searlier (1999) work exploring US employees' knowledge of rights regarding at-will-dismissal. Respondents were requiredt oi ndicate whethert hey felt the dismissalw as lawful or unlawful in respect of an umber of circumstances. But whilst incorrect responses were said to indicate aconfusion of legal normsand ethicalorsocial norms, correct responses wered eemed to indicate 'knowledge' of rights,r ather, than as may well have been the case,asituation in which ethical/social normsa nd the law happened to align (Kim 1999) or 'educated' or 'wild' guesses happened to be correct (Nadaeua nd Niemi1995). 5,6 Adifferent approach to exploring public knowledge of rights (whilst retaining afactbased questioning approach) was utilised afew yearslater by Barlow et al. in 2005. With the intent of exploring existing levels of knowledge among the general population as to the rights of ac ohabitant vis à vis the rights of as pouse, 7 the authors developed as eries of fact-based questions relating to the legal rights of aprotagonist in ahypothetical scenario (or 'vignette') which were then added to the 2004 British Social Attitudes Survey (BSAS). Finding that individualsh eld erroneous beliefs in respect of time-dependent accrual of benefits 8 and ab elief in the equivalence of paternal rights and responsibilities between married and unmarried couples, Barlow et al. (2005, p. 45) concluded that people's beliefs about cohabitation law, whether accurate or not, were based less on acquired knowledge, and more upon 'notions of social logic, fairness and morality'.
Whilst findings corroborated thoseo fo ther related studies (e.g. Meager et al.,2 002, Kim 1999) as Pleasencea nd  observed, the BSAS questions reliedo na presumption that peopleh ad good knowledge of the spousalr ights against which cohabitants' rights were beingc ompared, resulting in conflationb etweenk nowledge of spousalrights and knowledge of cohabiting rights. Thus, in their 2012 studyonthe same topic using respondents to the English and Welsh Civil and Social Justice Panel Survey (CSJPS), Pleasence and Balmer (2012) alsodeveloped aseries of fact-based questions in relation to ah ypothetical scenariob ut randomly varied the duration of the relationship (one month to 20 years -a llowing formal modelling of time dependency in beliefs about accrual of rights) and its status (cohabiting or married). Whilst employing ad ifferent methodology, like Barlow et al.,(2005), Pleasenceand Balmer (2012) found evidence of ongoing public misconception of cohabitation law, leading the authors to proposethat in the absence of an actual understandingo ft he law, the public tended to believe the law aligned with the rights they thought it ought to protect.
Although Pleasenceand Balmer (2012) were able to address some of the issues evident in the 'vignette' approach taken by Barlow et al.,(2005) this does not leave the approach limitation-free. Other signspoint to the fact that responses to hypothetical questions may, like self-reporting, provide only arough measure of knowledge for anumber of reasons. Nadaeu and Niemi(1995) note that someindividuals, even whenunsureofananswer,will be compelled to respond in amanner otherthan 'don't know' with somepeopleinclined to answer knowledge-based questions even in the face of considerable uncertainty (Chong 1993). For some, these uncertain answersare 'educated guesses',for others they may be 'wild guesses' (Nadaeu and Niemi1 995). In addition, it has been said that mostp eople 'construct' attitudeswhen asked aquestion (Tourangeauand Rasinski 1988), meaning that ar espondent without actual knowledge may answer af act-based questioni nm anner no different to how they would answer an attitudinal question; perhaps explaining why previous researchhas found that individuals respond with answersindicative of what they think the law should be, rather than whati ta ctually is. Theq uestiont hat these issues highlight, is whether there might be an alternative approach to exploring public knowledge of rights, and if so, what form it would take?

Alternative approaches to exploring public knowledge of rights
In the exploration of public knowledge of rights, open-ended questions appeart oh ave remained if not unused,then seemingly unreported. This is somewhat surprising since as Hruschka et al. ,( 2004) and Geer( 1988) note,t he use of open-ended questions offers a number of benefits, includingcapturing data of (potentially) greater accuracy or relevance to the individual, alongside capturing unanticipated responses.W hat may offset these benefits (and account for the lack of use to date) is as both Griffith et al. (1999) and Johnson et al.,(1974) note (although contested by Burchall and Marsh 1992) these benefits may comea tt he expenseo fh igher response rates, possibly because some individuals (more often those without educational qualifications according to Geer 1988) struggle to articulate their responses.
However, neither of theses uggested limitations would appeart oo utweigh the potential insight open-ended questions might provide whenused to explore how the public defines and describes their rights. This is particularly so in respect of knowledge of the law, since self-help and self-representationr ely (to varying degrees) on an individual's ability to identify and articulate an infringement of their rights.I nt he context of wideranging changestopublic legalaid in England and Wales and apresumedincreaseinselfhelp and self-representation (Public Bill Committee 2011), open-ended questions present a fresh approach to the measurementofp ublic knowledge of rights across ar ange of civil and social justice issues.
Accordingly, in this study we explore how individuals with one or more civil or social justice problemsr espond whena sked to briefly describet heir rights/legal position in the context of aq uantitative survey. Our findingsc ontribute to the existing methodological and PLE literature by: (a) examining the merits of usingopen-ended questionstoexplore public knowledge of rights; and (b) exploring how the public articulate their rights when asked to do so and what may be inferred about their level of knowledge. With the changes contained in the Legal Aid, Sentencing and Punishment of Offenders Act 2012 (LASPO) imminent, this study is particularly timely.

Aims and hypothesis
In thisstudy, we draw on data from the 2010 English and Welsh Civiland Social Justice Panel Survey (CSJPS) to build upon earlier studies exploring public knowledge of rights. Utilising self-assessment and open-ended (verbatim) responses from the CSJPS,this study sets out to determine: whether an open-ended question format can provide useful insight into an individual's knowledge of their rights and/or the validity of their self-assessment and the extent to which findings might provide useful lessons for research and policy development.
Referring to self-assessed knowledge of rights and on the basis of the issues raised in the extant literature (including the findingsofWilliams and Hall 1972, Caseborne et al. 2006, Genn et al.,2 006, Tennant et al. 2006 we hypothesise that the majority of those experiencing one or more civil or social justice problemsd on ot report having knowledge of their rights and that the majorityr eporta failure to acquire knowledge of their rights,but as found by Kim (1999) and Meager et al., (2002) theser ates will vary by problem-type. Based on Deming's (1950) concerns regarding issues of memoryrecall in relation to self-assessed questions, we also hypothesise that those who self-reportreport knowing their rights in Wave 1ofthe CSJPS in respect of a particularp roblem, frequently answer inconsistently whent hey are asked whethert hey knew their rights in Wave 2(in relation to the same problem). We further hypothesise that many of the people who profess to have knowledge of rights are unable to articulate these rights in response to the open-ended format, however unlike Geer (1988) we propose that the failure to articulate knowledge of rights will not be associated with education level.

Data
Data used for this studyc ame from the 2010 CSJPS, 9 an ationally representative survey of 3806 individuals aged 16 or over, living in 2318 households across 194 sample points in England and Wales. The survey explored respondents' experience of, and response to abroad range of problems involving legal rights. The surveywas conducted face-to-face in respondents' homes, with all members of households interviewed separately. The survey averaged 37 minutes, with ah ousehold response rate of 88% and ac umulative eligible adult response rate of 54%. For further technical details of the CSJPS, see Pleasence et al.,( 2011). Table 1s (Deming 1950, Groves 2009) mentioned above by distinguishingbetweenthose who knew their rights from the offset and thosew ho acquiredk nowledge. Those who stated that they knew their rights at the start of the problem 'completely' or 'mostly' and those who indicated that they now knew their rights 'completely' or 'mostly' were additionally asked whether they '[could] describe, briefly, what [your/your partner's] legal position was?' In Wave 2ofthe CSJPS, all follow-uprespondentswere asked again ' To whatextent do [you/they] now understand what[your/their] legalposition was?' in respect of the same problem,with the same range of responses available ('completely', 'mostly','partly', 'not at all', 'don'tknow'). This enabled the tracking of the self-assessment respondents gave, across CSJPS waves. However, owing to timing restrictions, follow-up respondentsi n Wave 2were not asked to briefly describe their rights,sowhere an individual responded that they knew their rights 'partly', 'not at all' or 'don'tknow' in Wave 1, butinWave 2 indicated that they now knew their rights,the verbatim responses of these individuals were not captured.

Analysis
Firstly we set out the extent to which respondents to the 2010 CSJPS (Wave 1) self-report knowledge of rights and how this varies with problem-type so as to provide an overview of the cohortand abasis for comparison with existing studies.Wedescribe the rate at which individuals knew their rights at the start of their problem and the rate at which individuals acquiredk nowledge of their rights,distinguished by problem-type. Exploring how self-reported knowledge changeso ver the courseo ft he reference period, usingd escriptives tatistics, we compare individuals' self-reportedk nowledge in Wave 1, against their self-reportedk nowledge in Wave 2, looking alsoa tt he extent to which knowledge varies by problem-type across Wave 1a nd Wave 2o fthe CSJPS.
Finally, we look at howindividuals answered the open-ended question asking them to describet heir rights. We code thesev erbatim responses into one of seven categories as outlined below. Whilst responses were not 'marked' as correct or incorrect, we anticipated that those inferring the intended meaningf rom the questiona nd capable of articulating their legal position/rights, would provide an answer that gave insight into their position of (legal) powerr elative to the other party. For example, someone who purchased af aulty coat and reported aconsumer problem could (in an idealscenario) answer the questionas follows: (a) 'underthe Sale of GoodsAct Iwas allowed to claim arefund or exchange on my faulty coat'. Those withoutk nowledge of or reference to the legislation might have offered as imilar explanations uch as: (b) 'the store had to providem ew ith ar efund or exchange'. In consideration of the fact that in the context of al ong quantitative survey some participants might have (reasonably) optedt oe xplain their legal position with greater brevity (see further Herzog and Bachman 1981), we could also have expected answerssuch as (c) 'I could get arefund' which although shorter,remained illustrative of an individual's knowledge of their rights/legal position.
Whilst the range of expected answerswas rather broad,individuals actually answered in afar more varied manner than this. Analysis of all verbatim responses identified seven different responses types that individuals tended towards in answering the question. These categories included: (1) Those who described their situationo rt he outcome they experienced (e.g. ' My zipper was broken on my coat' or 'The shop offered ar efund'); (2) Those who referred to legislation, used legal terminology, spoke of 'positive' rights,orprovided acommon sense interpretation of their rights (e.g. see (a), (b) and (c) above); (3) Those who referred to legislation,u sed legal terminology, or spoke of 'positive' rights but whoseresponse indicated that they may have misinterpretedtheir legal position, or over/underestimated their rights (e.g. an example of an overestimate of rights might have included: 'Under the Sale of GoodsAct my rights were to get a new coat and the shop had to replace it with abetter quality brand than previously and compensate me for inconvenience caused'. Whereas an underestimation of rights might have been: 'I was entitledt oarefund or return but it was the manufacturer In eeded to contact not the store'); (4) Those who madeav aluej udgement about the fairness of the legals ystem, the individuals involved,ortheir position in it (e.g. 'I was right, the coat was faulty'or 'No point in complaining, the Government is acting to take away the rights of consumers'); (5) Those who referred to broad rights that did not suggesta ny knowledge of their rights in relation to their problem,b ut were indicative of am ore general understandingo ft he legals ystem or with moral or ethical interpretations of the concept 'right'( e.g. 'HumanR ights', 'The right to voice my opinion', 'I could appeal', 'I had the right to be listened to'); (6) Those who claimed to have no legalposition/rights or to be 'in the wrong '(e.g. 'No rights', 'I should have noticed the zip was brokenbefore Itook the coat home, it was my fault Id idn't check'); (7) Those who claimed not to know, or who failed to articulate anything.
Whilst the answers provided weref or the mostp art discrete, where an individual provided ar esponse that couldh ave falleni nto two categories, they were assigned to a categorybased on the length of their answer. 10 This was not the case where an individual provided aresponse that may have fallen into category(2) and another category, they were coded only as (2) since the central issue was whethert hey could explain their legal/position rights, and not the information they provided surplus to this.
In exploring open-ended responses, we compare verbatim response-type against problem type,l ooking also at how individuals verbalised their rights, before turning to explore whetherresponse and articulation type varied by levelofeducational attainment.

Perceived knowledge and the acquisition of perceived knowledgeofr ights
Out of 1760 problems, respondents said they knew their rights at the outset of the problem, 'completely' for 396 problems (22.5%), 'mostly' for 319 problems( 18.1%), 'partly' for 356 problems (20.2%), 'not at all' for 612 problems( 34.8%). For af urther 68 problems (3.9%) 11 respondents suggested that they did not knoww hether or not they knew their rights.F or those1 036 problems where the respondent answered 'partly', 'not at all' or 'don't know', they were additionally asked whether they later came to understand their rights.F or 140 problems (13.6%)r espondentsc laimed to have acquiredk nowledge of their rights 'completely' in relation to the problem, for 205 problems respondentsacquired knowledge of their rights 'mostly' (19.9%), 'partially' for 345 problems (33.5%), for 289 problems( 28.1%) respondentsc laimed still to have obtained no knowledge and for 51 problemsrespondents claimed not to know (5%).
In total, discounting those who provided no response, for 715 problems( 41%) respondentsclaimed to know their rights completely or mostly at the start of the problem. For afurther 345 problems(20%)respondents came to acquire knowledgeoftheir rights, and for 685 problems (39%), respondentsd id not comet oa cquire knowledge of their rights. Table 2i dentifies theseindividuals on the basis of problem type.
As is showni nT able 2, thosew ith consumer problems more often knew their rights from the outset at 58.4% butw ere relatively less likely to acquire knowledge of their rights,f ollowed by ownedh ousing at 49.1% where the acquisition of knowledge was higher than non-acquisition. In domestic violence casesthe acquisition of knowledge was higher than its rate at the outset, butt hosew ith domestic violencep roblemsw ere more likely still, to not acquire knowledge at all, at ar ate of 37.8%. Af ailure to acquire knowledge was particularly prevalent for care proceedings, clinical negligence, education and neighbourscases. 12

Self-reported knowledge acrossC SJPS waves
Owingtothe panel format of the CSJPS, individuals at Wave 1were asked whetherthey knew their rights at the start of the problem or whether they later came to acquire knowledge. In Wave 2, follow-uprespondents(i.e. the same respondentsasWave 1) were again asked whether they 'now knew' their rights in relation to the same problem,allowing an exploration of whether respondents changedt heir self-assessment over time.R esults found that 92 respondents( 24.6%) claimed to have lost knowledge -t hat is to say they reportedknowing their rights to ahigherdegreeinWave 1, than they did in Wave 2. 135 respondents(36.1%) claimed the same level of knowledge, and 147 respondents (39.3%) claimed to have gained knowledge. Table 3displaysthe rate at which knowledge was lost, gained or remained the same by problem type.
As is shown in Table 3, those with neighbours (51.3%), personal injury (50%), employment(45.7%), clinical negligence (45.5%), and education (45.5%)problems were more likely to reportgainsinknowledge than those with otherproblem types. Those with owned housing (71.4%) and consumer (58.8%)were mostlikely to report the same levelof knowledge across waves, with thosew ith rented housing and personal injury problems reporting al oss in knowledge more frequently than other problems types. However, at a rate of 34.6% for rented housing and 33.3% for personal injury, approximatelytwo-thirds of individuals with thesesorts of problems reported the same, or again in knowledge. For all problem types, moreindividualsreported either gaining or retaining the same levelof knowledge than reported losing knowledge.

Open-ended articulation of rights
For the 1056 problems where individuals claimed to know their legal position/rights 'completely' or 'mostly' at either the outset of the problem or later during the course of the problem,i ndividuals were asked to briefly explain their legalp osition/legal rights. Analysis of the responses revealed some common themes categorised as described previously.  Table 4examines the categorisation of verbatim responses in relation to problem-type. As can be seen, 25.9% of verbatim responses were categorised as instances where the respondent described the situation or outcome of the dispute. Forty-two per cent of respondents detailed their legalp osition/rights referringt ot he legislation,l egal terminology or presenting al ay interpretation and fewer (8%)i ndividuals provided a more general/vague interpretation of their rights. Less frequently,i ndividualsg ave responses which suggested they may have under or overestimated their rights or the strength of their legal position at arateof2.6%, with 4.5% claiming that they had no rights in respect of their problem or that their 'legal'position was that they were to blame or at fault. 6.1% madeavalue judgement about the legal system in response to the question, with af urther 11.1% claiming that they did not knowo rc ould not verbalise their rights/legal position when asked.
Setting aside the categorisation of verbatim responses related to care proceedings given the small numbers contained within that group (fourintotal) resultsfurther highlight how individuals with certain problemst ended to reply in certain ways. This was not so evident in respect of those who responded to the open-ended questioni namanner that described their situationo ro utcome.H ere, relatively high rates of responding in this manner were seen across mostp roblem typesa nd differencesw ere relatively small with the exception of thosew ith education problems( at 12.1%). Where respondents did describetheir situation or outcome, their answers provided relatively little if any insight in to what they believedtheir rights to be, making it difficult to deduce whether they had any clear understanding of their rights as they believedt hey did. Exampleso ft hose who claimed they knew their legal position/rights included: 'Took out ajoint mortgage with a friend who had unknown to me apoor credit rating'and 'The agency gave me apoor score Ihad not done anything wrong they would not let me see my score in details'.Individuals with debt problemsfrequently spoke of the nature of their problem, rather than articulating what their legal entitlements were. Examples includedo ne who simply stated that they  (1) (3) ' went over their overdraft' and another who explained that 'I gave girlfriend moneyt o look after she spent it all.' In respect of those who did answer the question in the manner expected by describing their rights/legal position, theset ypes of responses were most common among those experiencing domestic violenceproblems at arate of 65%. Responses such as 'I could get him nicked' or 'I hadthe right to press charges' were typical of domestic violence cases. Those with consumer,rented housing,benefits, divorce/relationship breakdown,debt and education problemsalso had high rates of providing answersthat indicated that they had an understandingo ft heir legal position/rights at ar ate of between5 2-45%. However, whilst responses were indicative of somel evelo fk nowledge, the extent to which respondentsw eref amiliar with the legislation varied from individual to individual. For example, some respondentsp rovided relatively simplistic and short responses,s uch as 'had to pay the money'. Whilst this response presentsa ne fficient explanation in the context of al ong quantitative surveyi ti sd istinguished from thosew ho demonstrated willingness to provide am ore detailed response, by stating their obligations, along with their entitlements, e.g. 'had to repaythe money in affordable instalments' or 'I had to pay the money back at apayment that was affordable to me and not what they were asking'. Some individuals gave specificreference to the relevantlegislation in their answers with this more common in the case of thoser eporting clinical negligence or consumer problems. Fore xample, whilst some responses were again simplistic, e.g. 'could get a refund', others were more sophisticated, e.g. 'goods not sold as described so they did not meet Sale of GoodsAct', with another detailing that they had a'right to receive goods fit for purpose'. Although, individuals didn ot always recall the name of the legislation correctly,for example, one respondent claimed to be ' entitled to areplacement under the (non-existent) faulty goods act' this error did not compromisetheir understanding of their rights -n amely that they were entitledtoareplacement.
Yet, although frequently reporting their rights/legal position as expected,t hosew ith consumer problems alongw ith thosee xperiencing employment, rented housing and education problemsw ere alsom ore frequently categoriseda sr esponding to the openended questioninamanner that indicated they may have erred in their understanding of their rights, alongwith either potentially over or underestimating their entitlements (5.7%, 6%, 4.6%, 6.1% respectively). Generally, this overestimation appeared to derive from a misunderstanding of the relevantl egislation or itsa pplication. Some errors had the potential to impact upon the resolution of the problem and an individual's satisfaction with this resolution. Fore xample, one respondent claimed to have 'had the right to (have his vehicle repaired) to (his) satisfaction' wheninfact the relevantlegislation (SaleofGoods Act 1979) imposesa no bjective test of satisfactoryq uality (under s14 (2) and s48B as it relates to repair) suggesting the individual may have been waiting for an outcome which holds the other party to ahigherstandardthan the legislation does. There also appeared to be confusion surrounding tenancy rights, including one individual who erroneously believeditwas within their rights to 'withdrawrent until (their) property (was) sorted', a position which is not only legally incorrect, butp laces the individual at risk of eviction. Conversely, individualsalso appeared to occasionally underestimatetheir rights, including one individual who claimed that 'if you order something and pay on the Internetand tick the terms and conditions you have no rights' ap osition at odds with the entitlements enshrinedinthe Consumer Protection (DistanceSellingRegulations) 2000 and potentially also the Unfair Terms in Consumer Contracts Regulations 1999. 13 There was indication of potential underestimation of rights in respect of employment, with anumber of individuals claiming that they hadt op ut up with changest ot he terms and conditions of their employmento rg et another job. Whilst understanding that they did not have to agreet o such changes, they didnot appeartohave an appreciation of the fact that (legally speaking) getting another job was not their only alternative.
Value judgements were mostc ommon amongst those with neighbours (12.5%)a nd those with personal injury (19.6%) problems. Typically these were the typesofproblems where individuals provided very matter-of-fact and short responses to the question asked by detailing that they were' in the right', 'wholly innocent' or 'had done nothing wrong'. Whilst these were legitimate responses to the question asked, particularly since the question asked individuals to detail their 'legal position', they did not elicit the desired information, namely greater insight into whatp eoplet hought their rights were. It was also somewhat surprising to find responses discussing issues such as guilt/innocence in relation to acivil, rather than acriminal justice problem. However, this is perhapsless surprising given that personal injury problemstend to focus on the attribution of blame and responsibility, whilst neighboursproblems can often straddleboth the civil and criminal law.
In asimilar vein, those with employment and education problems more frequently gave answersindicative of them having 'no rights'withone individual claiming that they had no rights because they '(could not) argue against councilorgovernment decision'. At arate of 5.9% and 5.6% those with neighboursand benefits problems also expressedtheir belief that they had 'no rights' when asked. However, whilstsome appeared to be moreofapolitical statementthan an expression of legal understanding, for exampleone individual in respect of ab enefit problem who claimed that the 'government acting (was) to remove rights', other responses appeared illustrative, not of knowledge (or alack thereof) but whether a right couldr easonably be upheld by the individual. For example, one respondent with a neighbour problem claimed that 'essentially we had little rights, collecting evidence would be difficult and time consuming and we have to prove damage to property and then start civil action'. Similar such statements were evident in relation to employment problems where individualso ften claimed that they had 'no rights', not because none existed, but more often because they believedexercising such rights would puttheir job at risk.
Common sense interpretations of rights werem osto ften given in respect of clinical negligence, domestic violence, neighbours and owned housing. Examples includedthose who identified a' right to take an individual to court', 'the right to an appeal', 'human rights', or the 'right to be heard'or'voice an opinion'. Whilst these responses were again, not incorrect, they were indicative of am ore general understanding of an individual's position based on common sense or principlesoffairness, one not reliant on knowledge of the law in relation to ap articularp roblem. Of course manyp eople may also have had knowledge of the law in respect of the question asked -indeed one individual did make a distinction betweenw hats he saw as legal and moral rights in stating in respect of her employmentproblem that she 'Did not have legal right to move but had moral rights not to be movedf urther away from home'. However, because the rights these individuals professedt oh ave were so generic, it made it difficult to determine the extent to which these individuals had any clear understanding of the specific rights relative to their legal problem,o rw hethert hey were simplyr elying on ab road understanding of concepts of 'naturaljustice' or legal common sense.
Importantly, verbatim responses highlighted ad isjuncture betweena ni ndividual's belief that they knew their rights/legal position and their ability to articulate this. This was more so the case for thosereporting debt, benefits, education and owned housing problems than those reportingother problem types. For 117 problems (11.1%)individualsresponded that they '(didn't) know' when asked to articulate their rights,s uggesting that they were confidentenough to believe that their problem engagedaright but they couldnot explain this in legal terms.I ndeed, one respondent said as much, claiming that they '(couldn't) really describei t'. Table 5s hows the categorisation of verbatim responses in relation to the level of educational attainment of the respondents. Whilst there are some minor differences apparent in Table 5a ss hown below, there was no significant relationship emerging betweentype of verbatim response and levelofeducation qualification. 14

Discussion
Our results indicate that mosti ndividuals felt unsureo ft heir rights at the outset of their problem.F or 41% of problems, respondents claimed to know their rights completely or mostly at the start of the problem,f or af urther 20% of problemsr espondents acquired knowledge, and for 39% of problems respondentsdid notc ome to acquire knowledge.
Knowledge varied by problem type (see Table 2), with thosewith consumer problems more often reporting knowledge of their rights from the outset at 58.4% but relatively less likely to acquire knowledge of their rights, followed by owned housing at 49.1% where the acquisition of knowledge was higher than non-acquisition. In domestic violence casesthe acquisition of knowledge was highert han its rate at the outset, but those with domestic violencep roblems werem ore likely still, to not acquire knowledge at all, at ar ate of 37.8%.Afailure to acquire knowledge was particularly prevalent for clinical negligence, education and neighbour problems.
Utilising data from both Waves of the CSJPS highlighted that 24.6% of respondents claimed to lose knowledge with 39.3% claiming ag ain in knowledge. As is showni n Table 3, those with neighbours( 45.7%), personal injury (50%), employment( 45.7%), clinical negligence (45.5%), and education (45.5%) problemswere more likely to report gains in knowledge than thosew ith otherp roblem types. Those with ownedh ousing (71.4%) and consumer (58.8%)p roblemsw ere most likelyt or eportt he samel evel of knowledge across waves. Those with rented housing (34.6%)and personal injury (33.3%) problemsr eportedaloss in knowledge more frequently than those with other problem types.
Analysis of verbatim responses demonstrated an umber of response types, most common of these were responses in which individualswere able to give abrief overview of their rights/legal position. However, response typesv aried from those who supported their answer with reference to the relevant legislation to thosew ho simplys tated their rights.Inrespect of this latter group, although their answer appeared correct in relation to their dispute,i tw as not possible to deduce the basis of their knowledge, i.e. whether it stemmed from an understandingofthe legislation,orwhetheritwas based on acommon sense position as found by Kim (1999Kim ( ),M eager et al.,( 2002Kim ( ),B arlow et al. (2004 and Pleasencea nd . Individuals also frequently reported their situationo rt he outcome of the dispute rather than their legal/rights and tended to also provide very vague responses which did not appeart oh ave specific relevance to their particular dispute. In addition, whilstclaiming to know their rights,11.1% wentontoclaim they 'didn't know' in relation to the question. Less frequently, 2.6% of respondents gave verbatim responses which implied an over/undera ssessment of their rights. Results did not suggesta ny connection betweeni nterpretation of or response to the verbatim questiona nd levelo f educational attainment.

Who knows what and when?
In keeping with our first hypothesis, mosti ndividuals did not have knowledge of their rights to beginwith. These findingsare not surprising and are in line with the findingsof previous studies, notably that of Buck et al.,( 2008Buck et al.,( ),D envir et al.,( 2012,P arle/IARS (2009) and Casebourne et al. (2006). As also hypothesised, the majority (60.7%) of respondentswho reported alackofknowledge at the outset of the problem did not go on to acquire knowledgeoft heir rights to the standardof' completely' or 'mostly'. As found by Kim (1999) and Meager et al.,( 2002) and in keeping with our second hypothesis, we found that the rateo fe xisting knowledge and acquisition of knowledge varied by problem type. Interestingly, manyofthe problem types for which knowledge was poor at the outset, were related to social exclusion and vulnerability, including rented housing, money, debt, benefits, education and domestic violenceashas been noted by both Buck et al. (2005Buck et al. ( , 2008, Balmer et al.,( 2010) and Denvir et al.,( 2012).T his may be related to the complexity of the problem, or potentially the low rate at which individuals with these problem typess eek professional advice (see further Pleasence et al.,2 011,p. 43). This wouldalso explain why individualswith divorce and ownedhousing problems were more likely to acquire knowledge of their rights since these are problem typesf or which the obtaining of professionaladvice is far more common (Pleasence et al.'s 2011, p. 43).
In keeping witho ur thirdh ypothesis,b ased on Deming's (1950) and Groves et al.'s (2009) concerns regarding issues of memory recallinrelation to self-assessed questions, we found thatthose who self-reportedknowing their rights in Wave1of the CSJPS in respectofa particularproblem, frequently answeredinconsistently when asked whether they knew their rightsinWave2(in relationtothe sameproblem). However, more oftenpeople professed to having thesame if not greater levels of knowledge in relationtotheir probleminWave2. Thatreductionsinlevelsofknowledge happenedmore frequently forrentedhousing, money and personal injury problems may be ar eflection of the complexity of dealing withs uch issues or the fact thatindividualstended to overestimatetheir knowledge of rightsinrelation to these problem-typesmorefrequently,onlyrealising laterdownthe line.Thiswas certainly the casef or those withr entedh ousingp roblems where4 .6%o fv erbatim responses werec ategorised as '(3)P otentialE rror',t he third highesto fa ny other problem group. Interestingly,t hose withm oney problems tended towards answeringt he verbatim question by explaining their '(1)situation/outcome'thanthose withother problemtypes and those withpersonalinjury problems tended to report their rights as '(4)Value judgements' thano ther problemt ypes. The fact thatt heyd id not answer the verbatim questioni nt he expectedm anner and the factt hatt heya nsweredi ni taway thats uggested theyl acked knowledge of their rights, may account for why by Wave2theyh ad loweredt heir selfreported levels of knowledge.Whilst these resultshighlight how self-assessed knowledge changesover time, neithercomparisonbetween waves nor alteration of our question (toask participants about botht heir existing and acquiredk nowledge) can wholly negate the continued issue of memory-recallnoted by Deming (1950) and Groves et al.,(2009).
Consistentw ith our fourth hypothesist hat many of the people who profess to have knowledge of rights will be unabletoarticulate theserights in response to the open-ended format, we found that 11.1% of respondents were not able to articulate their rights/legal position whena sked to do so. However, an umber of other verbatim response types suggested that individuals had alack of understanding of their rights.Wewould contend that this was the case in respect of those58% of respondentswho when asked to explain their rights/legal position insteadgave asummary of their problem or its outcome, tended to over-or under-estimate their rights,g ave av alue judgement, mentioned non-specific rights or claimed to have no rights. Yet, although it suggests as much, we cannot conclusively state (contrary to their self-reporting) that these individuals did not know their specificrights in relation to the problem.Whilst on the face of it such responses imply if notalack of knowledge of rights/legal position, alackofunderstandingofwhatismeant by the term 'rights'/'legal position', these responses may simplyb ei ndicative of participants interpretingt he questiond ifferently than anticipated although we would not expectthis to be the case for the entire 58% of respondentswho did not answer the openended questionasa nticipated. What we did not findwas any evidence to supportGeer's (1988) assertion that response types would vary by education level. Although verbatim responses didvary, it did not appearthat education was associated with any clear patternof response or non-response as Geer(1988)

suggests.
What do open-ended questions tell us?
Our results supportt he proposition that an open-ended questiona pproach to exploring knowledge of rights yields insight into knowledge notacquired by other formats. Unlike hypothetical questions, asking individuals to describet heir rights/legal position in their own words gives insight into the knowledge driving their resolutionofthe problem. The fact that individualsw ho answered the fixed-choice self-assessment questiono ften then failed to answer the open-ended question( by claiming they 'don't know') highlights, as proposed by Schuman and Presser (1981) and Schwarz and Hippler (1991),the tendency of individuals to answer self-assessed/fixed-choice questions with greater confidence/frequencythan open-ended questions. However, the findings detailed above alsosuggest that the utilityofanopen-ended questionformat in the context of alarge-scalequantitative survey will remain constrainedb yd ifficulty reconciling articulation and interpretation with actual knowledge. Whilst this problem couldbeavoided through the use of vignettes it wouldb ed ifficult to do so with real problems. Although the open-ended question enabled individuals to answer freely,a sh as been previously noted ( Geer 1988) it also enabled them to freely interpret the question. There was also somee vidence,a s Tourangeau and Rasinski (1988) note in relation to fact-based questions, of individuals providing attitudinal responses rather than directly answering the question, as evidenced by those who provided avaluejudgment/claimed 'no rights' in response to the open-ended question. While nearly half of respondents were able to articulate their rights/legal position and interpreted the questionasanticipated, there was alimit to the extent to which responses provided clear insight.F or example, whilsts ome individuals referenced the relevantlegislation, most did not. Respondents may have been relyingon'common sense' when reporting what they believedtobetheir rights rather than an actual understanding of the law as found by Meager et al.,( 2002Meager et al.,( ),B arlow et al. (2004 and Kim (1999). Conversely, others may have provided an explanationo ft he situation, even thought hey did have aclear understanding of their rights.Thus, as found in respect of self-assessed, fact based,and hypothetical questions, an open-ended question is not in itselfcapable of providing conclusive insight into public knowledge of rights.

Policy implications
Studies seeking to acquire an understandingofthe extent to which the public know their rights,will tend to produceaninflated rate of knowledge where they rely on the public's self-assessment alone. As our results highlight, askingi ndividualsw hetherthey knew or acquiredknowledge of their rights yields agreater number of individuals saying yes than those capable of articulating these rights.S imilarly, approaches relying on tests of knowledge may just reflect the extent to which respondentsg uess correctly,r ather than reflecting any real knowledge. Including an open-ended question provides some degree of verification, however,i no rder to determine whethert he open-ended responses given derive from knowledge of the law, afurther question mustbeasked,designed to determine how the individual knew their rights. Based on both of the aforementioned considerations, am ore nuanced questiono rs eries of questions would need to deduce not only what individuals thought their rights to be, but alsoh ow they arriveda tt his conclusion. This may involve anumber of prompted open-ended questions that do not fit as readily within a quantitative surveyformat and may be more appropriatewithin the context of aqualitative study.
Whilst we would caution against concluding that most people were unable to articulate their rights simplyb ecause the answer they provided was not whatw as anticipated (particularly where this answer fellinto the '(1) situation/outcome' category) nonetheless our results suggest somep riorities for the development of public legale ducation. In particular, we note that individuals most often appeared unable to articulate their rights in respect of debt, benefits and education problems. Whilst this may be indicative of the complexity of the problem itselforthe capacity of the individual, if we are to assume that knowledge and articulation of knowledge of rights playsakey role in problem resolution, then public legal educationmay have arole to play in improving knowledge in these areas. In addition, individuals moref requently claimed to have 'no rights' in respect of employmentand education problemsthan in respect of otherproblem types. Ratherthan a misinterpretation of the question, such responses appear to be indicative of feelings of disempowerment that may stem, not from alack of knowledge of rights but rather asense of powerlessness arising from the inequality of the parties. Theperceived 'impotence' of these rights suggests the need to consider legislative rather than educational intervention/s. In contrast, those reporting 'vaguerights'(neighbours, owned housing, clinical negligence and domestic violence) may benefit from amore specific understanding of the problem as it relates to their rights. The same is also true of consumer, employment, education and rented housing disputes, where over/underestimation of rights occurred mostfrequently.
When considering family issues (divorce/relationship breakdown and domestic violence) in comparison to the remainingc ivil justice issues, individuals did tend to provide verbatim responses which were indicative of the fact that they knew and could articulate their rights.Ifthis (and the more frequent acquisition of knowledge for family law problems) is related to their obtaining of professional advice as noteda bove, any reduction in the availability of advice is likely to have an impact on the extent to which those with family law problems reportknowing their rights in the future. Given that family law is one area where legalaid is set to be substantiallycurtailed, it suggests an ongoing and long-term role for PLE in buffering the impact of legal aid reform.
Yet in spite of what appears to be continuedneed for PLE and whilstsimultaneously promoting self-help for the range of problems soon to be outofthe scope of legal aid, the Government has imposedu pon itselfn od uty to promotek nowledge of rights, develop just-in-time legal information, share the third sector's burdeno fe quippingc itizens to better handletheir problems alone or for that matter, inform itself as to the need for public legal education interventions (see further Legal Aid, Sentencing and Punishment of Offenders Act 2012, s1 (3) 15 ). Whilst an expectation that the public takesg reater responsibility for their legal problemss houldb ec omplimentedw ith the tools to enable them to do so, thus far it remains to be seen who will take responsibility for both the researchunderpinning and the development of PLE.

Future research
In this studyw eh ave used the articulation of rights as ap roxy for actual knowledge. Although we recognise that during the course of ad ispute somei ndividuals will never have to put into 'legal terms' the rights they feel have been infringed, our results provide new insight into how the public articulate their rights and whether open-ended questions might produce more reliable data. Nonetheless, further research utilising different methodologies is needed in order to morereliably determinehow well individuals know their rights and, crucially, the role knowledge playsi nt he resolutiono fc ivil and social justice problems. Our assumption remains that knowledge of rights leads to 'better' problem-resolution, but there is little reliable evidence to supportt his assumption and continuedd ifficulty in measuring outcomes. If we are to assume that individuals will in future be required to handle moreoftheir civil and social justice problems alone, it is clear that PLE will continue to have ar olet op lay in access to justice. But whethert his role should be to improveknowledge, build confidence, promote early action or acombination of theseobjectives, cannot be fully understood without further enquiry. i. The importance of the issue (including personal choices) ii. The litigant's ability to present their own case (including the venue before which the case is heard, the likely vulnerability of the litigant and the complexity of the law) iii. The availability of alternative sources of funding iv. The availability of alternative routes to resolving the issue v. The need to fulfil our domestic and international legal obligations (including those under the European Convention on Human Rights). 2. Civil Justice problems can also be referred to as 'justiciable', 'legal-related' or 'rights' problems, or 'problems with al egal dimension'. They are everyday and common-place problems that may or may not be resolved through the civil justice system. Examples of such problems are listed in Table 1. 3. PLE has been used to refer to both 'legal rights-based education' intended to promote general public awareness of the law and legal system and 'self-help/just-in-time' initiatives aimed at guiding people through the resolution of ap articular problem and the legal rights it engages (Giddings and Robertson 2003;Lawler et al,2 009). Rights-based education has included the introduction of compulsory citizenship classes within the school curriculum since 2005 and Ministry of Justice (MOJ) support of Public Legal Education (PLE) initiatives through the part funding of the Public Legal Education Network (PLEnet) (PLEAS Taskforce 2007). Efforts to provide self-help/just-in-time resources have not been centrally coordinated, but have included the publication of anumber of information leaflets available online and in person from arange of agencies including public and third sector organisations (e.g. Shelter, Citizens Advice, Money Advice Trust and government affiliated websites, including the Office of Fair Trading and the now defunct Community Legal Advice, DirectGov and National Debtline to name but a few). 4. This was particularly so in respect of discrimination-like actions which at the time did not fall under discrimination legislation, but which would have fallen under unfair dismissal, including dismissal on the grounds of sexual orientation and age. 5. For policy makers, this is perhaps not apressing concern where ethical/social norms continue to align, but does suggest that where the law changes, those people deemed to 'have knowledge' may not necessarily hold contemporaneous views any more. 6. In addition, one wonders whether the questions asked of Kim's (1999) respondents were those which individuals could reasonably have been expected to know. An example included a question which asked whether it was lawful to fire an employee as retaliation for refusing to participate in an illegal billing practice. Whilst the practice was lawful in New York (as opposed to the other two states where the study took place), for New York respondents to have known this would have required adetailed understanding of appeal court case law (Kim 1999, p. 469). Knowing the law in respect of rather obscure (and less likely) dismissal-at-will circumstances is arguably less of an imperative than providing ac orrect gauge of public knowledge relating to more likely dismissal-at-will events. 7. Theses ituationsi ncludedawoman'sr equest forfi nancials upport post-relationship breakdown, aw oman's entitlement to inheritance upon the death of her partner and an unmarried father's right to make decisions regarding medical treatment of his biological child vies-a-vie his rights as amarried man. 8. The idea that couples accrued greater rights as time went on. 9. The CSJPS is as ubstantially developed form of Genn's (1999) Landmark 'Paths to Justice' Survey. 10. So, where an individual provided ar esponse that could have been coded as (1) or (4) for example in relation to an employment problem they said 'I was in the right, Ihad been agood employee for over four years and Iw as then fired' they were coded as (1) because their explanation of the outcome of the case was longer than their (4) value judgement. 11. For nine problems respondents refused to give an answer (0.5%). 12. We also fitted two binary logistic regression models in order to predict the likelihood of respondents reporting knowledge of their rights from the outset (Model 1) or later gaining knowledge of their rights (Model 2) by problem-type whilst controlling for ar ange of sociodemographic characteristics (age, academic qualifications, gender, tenure, family type, illness/disability, mental health, ethnicity and income). Results yielded similar findings to those detailed in Table 2. Readers wishing to know more about these models should contact the authors. 13. For purchases made in England and Wales, with purchases from sellers within the EU protected by the European Union Distance Selling Directive 97/7/EC. 14. Even when splitting education level into abinary 'education qualification' and 'no educational qualifications' category, no significant pattern emerged to suggest that education level had any role to play in type of verbatim response given. 15. Note in particular, the wording used in Section 1(3) of the Legal Aid, Sentencing and Punishment of Offenders Act 2012, where it says 'The Lord Chancellor may secure the provision of-(a) general information about the law and the legal system, and (b) information about the availability of advice about, and assistance in connection with, the law and the legal system. This is ad eparture from the legal services envisaged under the Access to Justice Act 1999 which by virtue of s4 (1) stated that ' (a) the provision of general information about the law and legal system and the availability of legal services'w as responsibility of the Legal Services Commission and the Community Legal Service.