![Sample our Health and Social Care journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5938/TOC-Subject-Sample-2021-Health-Social-Care.jpg"})
Abstract
Over the last two decades the public sector has embraced new modes of service delivery, with a shift away from traditional face-to-face provision towards Internet and telephone based advice and information. While telephone provision has played an increasingly significant role in overall provision of legal advice in recent years, there has been limited empirical research which compares telephone to face-to-face services. Utilising administrative data from the Legal Services Commission (LSC) on legal aid services for housing problems, this paper explores the similarities and differences between the two delivery modes. We examine the client groups and matter types which tend toward particular channels of advice, the relationship between mode of advice and the outcome of cases for clients and the relationship between mode of advice and advice time. We find that there are significant differences in mode of advice among clients with particular demographic characteristics, with clients under the age of 18 and clients living with an illness or disability more likely to use face-to-face services. Our findings also suggest differences among the types of problems being addressed by telephone based services. On the surface there are small differences between modes of advice and the proportion of cases which lead to a substantive benefit outcome. However, once we disaggregate the data and investigate specific outcomes, there are pronounced differences in the outcomes achieved. Having controlled for key variables such as client demographics, case type and particularly stage reached, we find that telephone advice takes, on average, 14 minutes longer than face-to-face advice (compared to an hour less when examining the raw data). The implications of these findings for the future development of telephone based services in light of current policy in legal services are discussed.
Notes
1. The exceptions include the category of asylum, and other categories cases where the client needs emergency advice, where the client is a child, where the client is in detention or where they have been assessed by the CLA helpline as requiring face-to-face advice in the previous 12 months and are seeking to resolve a linked problem from the same provider. Exceptions would also be made for clients who would not be able to give or understand instructions given through a telephone-based service.
2. These include: the LSC's evaluation of an English Telephone Legal Advice Pilot in 2004; the Client Outcomes Hotlines Evaluation project which looked at seven telephone advice services in the United States in 2006 (Erlich et al. Citation2006) and the telephone hotlines effectiveness study, also conducted in the United States, which looked at five services in 2002 (Pearson and Davis Citation2002). It is noted that the latter refers to ‘Hotlines’ which are typically one-off requests for advice which differentiate them from the LSC model of provision of a full specialist help service delivered by telephone.
3. That is to say, only 23% specified that their preference was based entirely on mode, i.e. purely on the fact that they would rather obtain their advice over the telephone and not because telephone advice was more convenient on account of their distance from an advice centre, their inability to find another source of advice, their disability/illness, or their caring responsibilities.
4. The report also derived the conclusion of equal access on the basis that that percentage of BME clients utilising the service (13%) matched the rate at which they were represented in the general population, which does not necessarily follow.
5. Warm phones are located in Jobcentre Plus offices and allow clients to contact a number of other services free of charge.
6. The KPIs became formal contract requirements in the 2010 Standard Civil Contract but did not take effect until April 2012.
7. Controlled Work includes the following legal aid schemes: Legal Help which provides initial advice and assistance, Help at Court which allows someone to speak for clients at a court hearing, and Family Help which provides help with negotiation and obtaining court orders in family disputes. It does not include Legal Representation for clients taking or defending court proceedings.
8. Civil legal aid work for the year 09/10 was conducted under the Unified Contract, which replaced the LSC General Civil Contract and Family Mediation Contract in April 2007. This contract brought conditions for Not for Profit organisations into line with solicitors who carry out civil legal aid work. From 1 October 2007, Not for Profit providers were covered by the same payment provisions as solicitors, subject to some transitional provisions.
9. Matter Type I categories are: ‘Homelessness/threat of homelessness’; ‘ASBOs – Magistrates/Crown Court’; ‘Housing benefit’; ‘Anti-social behaviour not involving criminal courts’ (referred to as simply ‘Anti-social behaviour’ in the text); ‘Landlord and tenant: other terms and conditions’ (‘Landlord and tenant’ in the text); ‘Possession – mortgage’; ‘Other’; ‘Possession – other’; ‘Re-housing (non-homelessness)’ (‘Re-housing’ in the text); ‘Disrepair’; ‘Possession – rent arrears’; ‘Harassment/unlawful eviction’. Matter Type II categories are: ‘Client has public landlord’; ‘Client has private landlord; ‘Client has other social landlord’; ‘Client has NASS accommodation’; ‘Client is owner occupier’; ‘Client is homeless’; ‘Client is landlord’; ‘Other’.
10. The stages are defined as: First Meeting, where no further work is undertaken beyond confirming instructions and giving advice, including the confirmation of advice letter; Further Work: Work beyond the first meeting, including work preparatory to putting the case for the client. This applies to where further work has been carried out but there has been no contact with the potential opposing party, other than for the purposes of investigation; Putting the Case for the Client, Includes correspondence with opposing party and pre-action protocol letters and any communication with the third party that involves substantively putting the case for the client, including preparation of representations and evidence for tribunals; Representation at Court/tribunal, applies where representation before courts or tribunals has been provided or arranged.
11. In the housing category, the following outcomes are recognised as providing a substantive benefit: Client receives damages or property, Client receives new or increased periodical payment, Client receives damages or property and new or increased periodical payment, Sum owed by client to a third party is reduced or is less than claimed, Liability of client to make regular payments is reduced or is less than claimed, Client housed, re-housed or retains home, Repairs or improvements to the client's home, Opponent/other party action benefits client, Opponent/other party action prevented, Opponent/other party action delayed, Client secures explanation or apology, Client advised and enabled to plan and/or manage their affairs better. Outcomes which are not defined as providing a substantive benefit are: matter concluded otherwise, matter stopped on advisor's recommendation, matter proceeded under other CLS funding, client referred to another organisation, client advised and taking action themselves or with the help of a third party, client advised and third party action or decision awaited, and outcome not known/client ceased to give instructions.
12. Note, a multilevel model could also be fitted here, since cases are nested within supplier. Using a single-level model assumes that clients make the choice to use the telephone or seek face-to-face advice independently of any choice of adviser (which is likely to be the case for the majority of clients). If clients were to choose advisers, with mode then driven by advisers, a multilevel model would be more appropriate (which was considered less likely). Readers interested in corresponding multilevel findings should contact the authors.
13. Note, that stage reached (First Meeting; Further Work; Putting the Case for the Client; Representation at Court/tribunal) was not included in the ‘mode of contact’ analysis. This decision was taken since the assumption was made that mode was chosen independently of any consideration regarding the stage that was likely to be reached upon the conclusion of the problem (which would only become clear well after choosing a mode and as a consequence of the advice received). However, if stage was considered a proxy for severity/intractability, and clients were making decisions on mode in light of this, its inclusion may be justified. As a consequence, we also refer to findings with stage reached included as endnotes throughout the ‘mode of contact’ analysis. Readers interested in the full statistical output having included stage should contact the authors.
14. The impact of age remains if stage is added to the model. Parameter estimates (and standard errors) were 0.61 (0.11) for 18- to 24-year olds, 0.57 (0.11) for 25- to 34-year olds, 0.51 (0.11) for 34- to 44-year olds, 0.52 (0.11) for 45- to 54-year olds, 0.57 (0.11) for 55- to 64-year olds, 0.53 (0.12) for 65- to 74-year olds and 0.79 (0.14) for those aged 75 or over.
15. These difference remained having added stage to the model, though their significance, particularly for ‘other’ ethnicity was reduced (χ21 = 85.55, p < 0.001 for ‘mixed’ and χ21 = 6.31, p = 0.01 for ‘other’).
16. While some reduction in significance was observed, all of these terms remained highly influential having added stage to the model (χ21 = 46.27, p < 0.001 for physical ill-health; χ21 = 336.67, p < 0.001 for mental illness and χ21 = 22.97, p < 0.001 for ‘other/unknown’).
17. Findings for matter type 1 were predominantly similar having introduced stage to the model (for example, χ21 = 111.33, p < 0.001 for housing benefit; χ21 = 1621.85, p < 0.001 for landlord and tenant issues and χ21 = 37.69, p < 0.001 for ‘ASBOs in the magistrates or crown court’).
18. Again, findings for matter type 2 were fairly comparable having added stage to the model (for example, χ21 = 91.41, p < 0.001 for ‘other’ and χ21 = 201.85, p < 0.001 for owner/occupiers).
19. Particularly when using large datasets, statistical significance will not equate to practical significance. This highlights the importance of also interpreting models in terms of actual percentages, probabilities or times. For example, the statistically significant mental illness term in this model equates to 67.4% substantive benefit compared to 68.7% for the reference category (simulated from the model), a modest difference in practical terms.
20. Providers of specialist telephone advice are permitted to provide up to 2 hours of advice before evidence on legal aid eligibility must be submitted by the client.