Writing reports in social work: Characteristics,correctness and style

Introduction

Since the 1990s, the field of social care has experienced an increase in bureaucratic procedures, linked to the rise of managerialism, typical of the neoliberal economic context, where results and evaluation indicators have become particularly important (Fernández and Jiménez, 2006: 19).

Of all the tasks performed by social workers, writing is perhaps the most time-consuming. As Rai (2014: 4, based on Holmes et al., 2009; O’Rourke, 2010) has pointed out, ‘writing takes up more than half of social workers’ time’.

In their daily practice, social workers use writing to codify, analyse, interpret and transmit the information they receive from clients, as well as to share assessments, make recommendations and even use it as a tool for intervention and therapeutic supervision (Rai, 2014: 4; Timms, 1972: 23).

Social workers write a wide variety of documents, including case records, basic assessments, placement information, observation records, court statements or reports (Rai, 2014: 111).

Even to write the simplest document, social workers must draw on a wide range of academic and professional skills. These include, among others, selecting information and conveying it clearly and unambiguously (McDonald et al., 2015: 371).

Together with case recording, reports are the main tool used by social workers to describe the situation of service-users, providing evidence of their needs and supporting proposed actions (Fernández and Jiménez, 2006: 22; Giribuela and Nieto, 2010: 17; Rai, 2014: 112; Roose et al., 2009). In this sense, Rai (2014) defines the report as ‘a structured piece of factual writing that is written for a specific purpose and usually offers some sort of recommendation’ (p. 166).

According to this definition, reports would fulfil two fundamental objectives (Colom, 2005: 123; McDonald et al., 2015; Munuera, 2012: 153):

However, meeting these two objectives (especially the second one) may involve ethical dilemmas for social workers, related to the duty to inform third parties and respect for confidentiality (Ballesteros et al., 2012; Banks, 1997; Munhoz and De Souza, 2017). Social workers are responsible for defining the purpose of the report, selecting its information and choosing its receivers carefully (Rai, 2014: 133). As Clapton (2018: 5) states, they need to be aware that the use of language can ‘shape meanings and actions’ and that a report can bring about change in the lives of the people to whom it refers (Heffernan, 2006; McDonald et al., 2015: 360).

These ethical dilemmas are related to the conception of the report as a document that can have an impact on the life of the person to whom it refers. As Giribuela and Nieto (2010) point out, social workers ‘report for something, and that something must be clearly stated in the body of the report’ (p. 20). If this is not the case, the purpose for which it was written may not be achieved, which could mean the failure of the intervention. This is one of the reasons why having a set of useful writing tools for writing reports can improve the social worker’s intervention, helping to promote the best interests of the service user in the context of available resources (McDonald et al., 2015: 363–364; Rai, 2014: 133).

Social organisations are increasingly demanding succinct, effective, objective and reliable reports (Kim et al., 2021; Munro and Hardie, 2019; Sharpe et al., 2011 in Nelson and Weatherald, 2014). However, the social work literature does not explain how to achieve this (Roscoe, 2014). Professional writing is a new research topic in social work and is less valued than other forms of communication such as oral skills (Masocha, 2017; Healy and Mulholland, 2007 in Nelson and Weatherald, 2014). Thus, according to Giribuela and Nieto (2010: 7), ‘we write reports, but we do not write about reports’, and the professional literature on writing social work reports is clearly insufficient (Fernández and Jiménez, 2006: 17; Rai and Lillis, 2013: 353). Reports belong to a discursive genre with descriptive and evaluative functions. Through reports, social workers construct evidence-based arguments that, in general, should offer evaluative judgements about people and their situations (Rai, 2014: 167).

To fulfil this role, the following skills are desirable: ability to select information appropriately, ability to use evidence, ability to write concisely and ability to analyse and interpret information (McDonald et al., 2015; Rai, 2014; Timms, 1972). Ultimately, it is about the social worker being able to write effective reports.

In 1967, the philosopher of language Paul Grice gave his famous lecture ‘Logic and Conversation’ (included in Studies in the Way of Words, 1975) in which he reflected on how the effect of speakers’ behaviour should be to render communication optimal. For him, optimal communication (where the receiver interprets exactly what the sender wants to express) can only be achieved if the communicative act fulfils four maxims or pragmatic conditions, known as Grice’s maxims (Grice, 1975, in Escandell, 2020: 81; Reyes, 2021: 39). Grice’s maxims are as follows: Manner (clarity), Quantity (conciseness), Relation (relevance) and Quality (reliability).

Thus, to carry out this research, we start from the theoretical basis of correctness of language offered by Grice (1975), linking his four maxims with the recommendations on writing and style identified in the professional literature on social work.

Therefore, the main objective was to analyse the content and writing style of the reports written by social workers in a Social Care Centre (SCC). The specific objectives were: to learn about the experiences of social workers in report writing; to analyse what criteria of linguistic, grammatical and stylistic usage should be observed in social report writing; to analyse a sample of reports, taking into account criteria of clarity, conciseness, relevance and reliability; and to identify factors that facilitate report writing and improve the quality of reports.

Method

The data were collected in an SCC located in Madrid (Spain). This SCC is part of the Spanish network of primary SCCs of councils, which are the gateway to social coverage for citizens. The SCCs prevent and treat social exclusion situations through interdisciplinary interventions. They offer guidance and advice to citizens, assess their needs and determine the benefits and services to which they can have access. The attention offered aims to contribute to the social welfare of all citizens, without making distinctions by groups or specific needs. If appropriate, the person is referred from primary care to the second level of care, the specialised social services.

The SCC under investigation serves a population of approximately 35,000 people (INE, 2022). The profile of the people attended to is very varied, since primary care has a multipurpose nature and channels all kinds of requests and needs, within its competencies and framework of action. It offers three types of services:

The research was carried out in the Social Work Unit of this SCC and consisted of two stages. The first involved semi-structured interviews with social workers from the Social Work Unit at the SCC. The second entailed the evaluation of a sample of reports written by social workers from the SCC. These two stages are described below, following the same structure: participants or units of analysis, instruments and procedure.

Instrument for classifying the reports

In 1985, Spain’s General Council of Social Work (Consejo General de Colegios Oficiales de Diplomados en Trabajo Social y Asistentes Sociales, 1985) established the structure and minimum content of the reports to be signed by social workers. This criterion was modified in 2005 by the publication of the Green Book of Social Work, written by Dolors Colom Masfret (2005), also published by Spain’s General Council of Social Work. This book established a new series of variables for the writing of reports, still in force today.

Since the difference in criteria between the two manuals is substantial (the first of them only includes 7 criteria, while the second reference includes a total of 20), two templates were designed:

Instrument to analyse reports’ writing style

After reviewing the literature in social work, it was noted that references to writing techniques or skills often do not offer comprehensive or organised criteria, which undoubtedly makes their application difficult.

For this reason, we turned to other disciplines, such as linguistics, with the aim of trying to identify and classify the most important variables about writing style. From among the existing theories, Grice’s theory (1975) was selected for two main reasons:

The content and writing style of the 250 reports were evaluated using a template based on the four pragmatic criteria known as Grice’s maxims (1975, in Escandell, 2020: 81). These maxims are as follows: Manner (clarity), Quantity (conciseness), Relation (relevance) and Quality (reliability).

These maxims were compared with the writing style indicators found in the specialised social work literature. Indicators with the highest prevalence and greatest agreement were selected and categorised as follows:

These indicators were included in Supplementary Table S3, designed for this purpose, which is included as supplementary material.

The evaluation was carried out by two persons:

To calculate the degree of agreement between reviewers, it was decided to use Cohen’s kappa coefficient, which takes into account the probability of agreement due to chance. Table 1 shows the coefficient for each of the indicators. As can be seen, all are high and none is less than 0.30, which implies a correct degree of agreement. The mean of the coefficients was 0.65, which can be considered a strong or substantial degree of agreement and a general indicator of the validity of the instrument (Landis and Koch, 1977).

OPEN IN VIEWER

Table 1.

Inter-rater agreement for each of the style correctness indicators measured by Cohen’s kappa obtained from a 10 percent sample of the total 233^a reports (n = 23).

Indicators	Cohen’s kappa coefficients
1. Widely used vocabulary	1
2. Functional vocabulary	0.65
3. No spelling mistakes	0.30
4. Relevant professional vocabulary	0.47
5. Clear structure divided into paragraphs	1
6. Simple sentences	0.47
7. Avoids repetition	0.82
8. Reformulation as technique for clarification	0.65
9. Avoids disconnected fragments	0.82
10. Descriptive, explanatory concreteness	0.30
11. Expressions and terms with a high semantic content	0.47
12. Avoids vagueness	0.30
13. Avoids stereotyped expressions	0.82
14. Correct use of person in verbs	1
15. Sources explicitly referenced	0.82
16. Different versions provided when several parties are involved	0.82
17. Assessment of only what is relevant to report	0.47
18. Avoids any kind of stereotypes, clichés and prejudices	0.30
19. Information relevant to report subject	0.82
20. Avoids anecdotes	0.65

Each report was scored for each indicator using a three-item scale (fully consolidated, consolidated and unconsolidated) and Yes/No responses.

Several indicators were established for each variable. One point was awarded for each indicator.

The results make it possible to assess whether the presence of the variable analysed is not well consolidated, sufficient (consolidated) or insufficient (not consolidated).

Supplementary Table S4, prepared for this purpose, is included as supplementary material.

Results and discussion

Below are presented the results obtained after applying each of the techniques mentioned (semi-structured interviews with the social workers; content of the social reports that make up the sample; and analysis of the writing style of these social reports).

Interview results

The content of the interviews was analysed by the authors. First, the interviews were transcribed, and the information was sorted according to the subject areas. Second, the responses were grouped according to their similarity. The following are the highlights of the main topics discussed during the interviews with social workers:

Analysis of the sample of reports

Minimum content of a ‘social report’

Reports dated before 2005

Personal information is included in 100 percent of these reports. Similarly, the person or family subject of the intervention is clearly identified in all reports. Information on ‘Socio-economic status’ (86.3%) and ‘Family unit characteristics’ (82.3%) is present in most of the 51 reports dated between 1985 and 2005.

Information on ‘Interpretation and assessment of the situation’ was found in 80.4 percent of these reports. Core assessment refers to the professional opinion issued after studying the situation, which must incorporate an analysis and interpretation of the information (Del Caño et al., 2012). However, in some reports, the content of this assessment is insufficient, including only standard phrases, such as ‘In view of the above . . .’, followed by a proposal for action. In these cases, core assessment in indeed present and has been considered in the percentage of results, but its truth also lacks content and does not fulfil its intended purpose.

Since most of the reports are related to obtaining benefits and payments, it is understandable that they contain information about the family unit and its socio-economic situation. Other sections such as ‘Social and neighbourhood relations’ only appear in 5.88 percent of the 51 reports dated before 2005.

‘Household health information’ is only found in 52.95 percent of the reports, under this or another name, while ‘Social environment characteristics’ appears in just 45.09 percent. According to Del Caño et al. (2012), this may be because the report’s objectives determine the variables found in the body of the report. In this regard, it is important to recall that in Spain, according to the General Council of Social Work (Consejo General de Colegios Oficiales de Diplomados en Trabajo Social y Asistentes Sociales, 1985), all social reports issued before 2005 were required to include this information, regardless of the purpose for which they were issued. The results, however, highlight the possible gap between institutional or theoretical indications and professional practice.

Reports dated after 2005

Among the 199 reports issued between 2005 and 2015, the most frequent content was ‘Personal information’ (99.49%, 198 of 199 reports). Along the same lines, all but one of the reports have explicit client/family unit information.

As mentioned above, currently in Spain, the Green Book on Social Work (Colom, 2005) establishes the minimum content valid for any report issued by a social worker, regardless of the type, agency or purpose of the document.

The Green Book requires a report to include a total of 20 variables. However, none of the 199 reports analysed fully complied with this requirement. Most reports contain between 14 and 16 of the variables. This leads one to wonder whether the list of variables required by the Green Book (Colom, 2005) is too exhaustive, especially considering that one of the main ethical dilemmas faced by professionals, according to the study by Ballesteros et al. (2012), is the appropriateness of sharing or disclosing information to other professionals. In turn, the data again seem to suggest a possible gap between theory and professional practice, as it seems difficult to reach a consensus on the content of the report in a profession such as social work, which is characterised by working in plural contexts and whose intervention varies depending on the group of users and the agency.

The analysis of this set of 199 reports shows that some of the variables were often present in the document body rather than under the heading or subheading recommended by the Green Book. These contents appeared under other headings, mixed in with the rest of the text. This happened especially with other interrelated criteria alluding to the family unit’s financial situation: ‘Employment status’ (69.84%), ‘Socio-economic status’ (68.84%) and ‘Housing characteristics’ (56.28%).

The dimensions ‘Issued by’ and ‘Date of issue’ were found in 97.98 percent of the documents reviewed (195 of 199). ‘Issuing institution’ was present in 97.48 percent of the reports and ‘Receiving institution’ in 84.92 percent.

Following Del Caño et al. (2012) and Rai (2014), it can be argued that these percentages are significant because including an issue date helps emphasise that the report has an expiry date, has been prepared for a specific moment in the client’s life and cannot be extrapolated to other purposes or times. In addition, the fact that the social worker and the institution issuing the report are identified in a large number of the reports helps create a group identity and gives a more professional appearance.

‘Core assessment’ (81.40%) and ‘Technical recommendation’ (70.85%) criteria were found extensively. However, interpretation of the information was missing since most reports simply summarise information about the service-users’ life histories.

In the section of variables with less representation in the social reports analysed, the following groups were found:

‘Demand made by the person’: data that only appear in 35.67 percent of the social reports reviewed.

‘Relationships’: variable present in 28.64 percent of the documents analysed.

‘Positive aspects of the person/family and environment’: only specified in 12.06 percent of the social reports collected.

‘Subjective perception of the person about his or her situation’: these data appear in 10.05 percent of the documents reviewed.

These four variables refer to the person’s perception of his or her situation, their empowerment and the strengths and resources that he or she can mobilise to change it. These are data that allude to the proactive capacity of the user rather than considering him or her as a mere passive agent who is a beneficiary of resources and benefits. However, their low representation in the sample of 199 social reports analysed suggests that these dimensions are not yet taken into account by professionals, or that they are not considered worthy of mention when drafting a social report, a circumstance that should be corrected in order to focus attention on users’ empowerment and not on their shortcomings or needs, as stated by Fernández and Jiménez (2006), Del Caño et al. (2012) and Rai (2014).

Writing style of a sample reports

Of all the reports, 17 did not have sufficient content to be analysed. These 17 reports were linked to a standardised report template for accessing the benefits provided under Law 39/2006 on the Promotion of Personal Autonomy and Care for Dependent Persons.

As shown below, the remaining 233 reports showed better indicators for clarity, relevance and reliability than conciseness, a criterion that was repeatedly violated.

Clarity: ‘Achieved’ in 127 reports (54.50%) and ‘well achieved’ in 54 (23.17%). Thus, 77.68 percent of the reports had explanations of whose clarity was acceptable or good.

Relevance: 59.22 percent of the reports were classified as ‘well achieved’ (138 of 233) and 27.46 percent (64 of 233) as ‘achieved’. This gives 86.68 percent of the reports that were relevant.

Reliability: 52.78 percent of the reports were classified as ‘not achieved’ (123 of 233). However, this figure was partly balanced by the reports classified as ‘achieved’ or ‘well achieved’, with an overall result of 47.22 percent of social work reports with reliable content.

Conciseness: This maxim was found to be ‘not achieved’ in 69.52 percent (162 of 233) of the reports. This is a remarkably high figure compared to the 18.88 percent of the total (44 of 233) whose conciseness was ‘acceptable’ and the 11.58 percent (27 of 233) for which it was ‘good’.

Table 2 and Figure 1 show a breakdown of the data.

OPEN IN VIEWER

Figure 1.

Number of reports with score not achieved (black), achieved (dark grey) or well achieved (grey) on each of Grice’s maxims (n = 233).

OPEN IN VIEWER

Table 2.

Number and relative frequency of reports according to the degree to which they have reached each of Grice’s 4 maxims (n = 233).

Maxims	Not achieved	Achieved	Well achieved
Clarity	52 (22.31%)	127 (54.50%)	54 (23.17%)
Conciseness	162 (69.52%)	44 (18.88%)	27 (11.58%)
Reliability	123 (52.78%)	104 (44.63%)	6 (2.57%)
Relevance	31 (13.30%)	64 (27.46%)	138 (59.22%)

The highest values are shown in bold.

Indicators of clarity

The ‘Widely used vocabulary’ indicator was found in 98.28 percent (229 of 233) of the reports. So, most of the texts were written with widely used vocabulary, typical of standard educated usage, without jargon or obscure vocabulary.

This, together with a ‘Clear structure divided into paragraphs’ (82.83%), ‘No spelling mistakes’ (68.66%) and ‘Simple sentences’ (45.49%), indicates that most of the reports achieve the clarity maxim.

The 31.34 percent (73 of 233) with spelling mistakes seems a relatively high figure considering that social workers must have a university degree. However, it should be taken into account that these spelling mistakes also include any inappropriate use of commas or other punctuation marks and accents, which somewhat mitigates the result.

‘Relevant professional vocabulary’ was used correctly in 41.20 percent of the reports (96 of 233), while ‘Functional vocabulary’ was only found in 21.45 percent (50 of 233).

Figure 2 shows a breakdown of the data.

OPEN IN VIEWER

Figure 2.

Number of reports that meet (black) and do not meet (grey) each of the identified indicators within the Clarity maxim (n = 233).

Indicators of conciseness

In the analysis of this maxim, the result for the ‘Expressions and terms with a high semantic content’ indicator stands out. It was not found in 83.27 percent of the reports (194 of 233). Groups of three or more words were used to explain something that could be synthesised in a single term, making it difficult to be concise. One possible explanation may be offered by Rai (2014), who mentions that ‘people have very different understandings and expectations about what “clarity” and “conciseness” are’ (p. 169).

Of all the reports, 77.69 percent (181 of 233) were considered vague and imprecise and were full of ancillary clarifications that made them difficult to read. Some 63.52 percent of the reports (148 of 233 documents) included ‘Stereotyped expressions’, with an implicit risk of perpetuating prejudices and clichés. The ‘Avoids repetition’ indicator was found in 62.23 percent of the reports (145 of 233), which means that the narrative was not repetitive. However, 59.66 percent of the documents reviewed (139 of 233 reports) lacked ‘Descriptive, explanatory concreteness’. This means that the essence of the discourse was lost in circumlocutions and overuse of the passive voice.

‘Avoids disconnected fragments’ was found in 59.22 percent (138 of 233) of the reports, which improved the conciseness of the overall text.

Finally, only 24.46 percent of the reports (57 of 233) used ‘Reformulation as technique for clarification’ correctly. This outweighs the greater recapitulation of arguments, leading to reiteration and increasing the percentage for this indicator.

Figure 3 shows a breakdown of the data.

OPEN IN VIEWER

Figure 3.

Number of reports that meet (black) and do not meet (grey) each of the identified indicators within the Conciseness maxim (n = 233).

Indicators of reliability

In this category, the most relevant result was that 96.56 percent of the reports comply with the ‘Correct use of person in verbs’ (225 of 233).

The ‘Assessment of only what is relevant to report’ and ‘Avoids any kind of stereotypes, clichés and prejudices’ indicators were met by 51.93 and 53.21 percent of the reports. This implies that 46.79 percent of the documents did contain stereotypes, clichés, or prejudices, which seems a very large proportion given that this compromises the reliability of the reports.

In addition, only 9.01 percent of the reports (21 of 233) cited the sources from which the information had been obtained. Only 15.02 percent included different versions when several actors were involved (35 of 233 reports), and this gives an indication of the low credibility and rigour of the documents analysed (Del Caño et al., 2012; Rai, 2014: 168).

Figure 4 shows a breakdown of the data.

OPEN IN VIEWER

Figure 4.

Number of reports that meet (black) and do not meet (grey) each of the identified indicators within the Reliability maxim (n = 233).

Indicators for relevance

Finally, it is notable that 59.22 percent of the reports contained information relevant to the report’s purpose (138 of 233). This result is similar to that obtained in the survey by Del Caño et al. (2012). They concluded that only 57.89 percent of the reports contained relevant language and information.

A positive figure is that 84.54 percent of the reports reviewed (197 of 233) did not contain anecdotes or superfluous information, making them relevant and appropriate to the report’s purpose.

Figure 5 shows a breakdown of the data.

OPEN IN VIEWER

Figure 5.

Number of reports that meet (black) and do not meet (grey) each of the identified indicators within the Relevance maxim (n = 233).

Conclusion

Before moving to the conclusions, it is important to note that the results of our study are limited to their specific context (Spanish) and to social workers like those who kindly participated in the research.

The interviews with the social workers at the SCC and the analysis of the sample of 250 reports showed two fundamental and similar results:

The analysis of 250 reports produced a remarkable result: none met the minimum criteria to be considered a proper report. This was true when looking both at those criteria recommended by the Manual of the General Council of Social Work (Consejo General de Colegios Oficiales de Diplomados en Trabajo Social y Asistentes Sociales, 1985) and those suggested in the Green Book (Colom, 2005). Although both manuals underline that there are circumstances under which reports do not need to fulfil the criteria for all the variables, in general, reports should fulfil them all.

The academy advocates an open model for reports, one which can be adapted to suit each case and includes variables strictly related to each report’s purpose (Del Caño et al., 2012; Fernández and Jiménez, 2006; Hernández, 2003; Munuera, 2012; Rai, 2014).

However, some of those interviewed suggested that there is a need for a new professional instrument that, while less exhaustive than reports, could be helpful for dealing with requests for specific information. This would not simply be a report with fewer variables issued under a different name. Instead, it would involve designing a new instrument meeting the information needs of government departments, organisations and other bodies without jeopardising or degrading the standard report.

It is important to note that our analysis of how reports fulfil the maxims proposed by Grice in 1975 (in Escandell, 2020) shows that reports are notable for their clarity and relevance. However, their writing style needs substantial improvement regarding conciseness and reliability to increase their effectiveness as professional communication tools.

Concise writing is essential for reports to achieve their purposes and ensure that their recipients can understand them. Reports must contain all the information needed to get the message across but no more (McDonald et al., 2015; Rai, 2014; Roose et al., 2009).

The specialist literature concurs that a concise writing style helps to improve the perception of reports as a social work tool because it provides rigour and protects information confidentiality while maintaining professional ethical standards (Ballesteros et al., 2012; Del Caño et al., 2012; Heffernan, 2006; Munro and Hardie, 2019; Roose et al., 2009: 328).

However, as well as being concise, reports must be reliable. Reliability is based on truthful and impartial information. This is essential to ensure that reports comply with the ethical precepts in the Social Work Code of Ethics (McDonald et al., 2015: 365; Munro and Hardie, 2019; Roose et al., 2009).

Sometimes the report written by the social worker will include a recommendation or a conclusion about the person’s life; namely, it will be a document with a persuasive character. However, persuasive does not mean manipulative. To ensure that this persuasive writing remains within professional ethics, it has to fulfil two main qualities: to be truthful and to include the voice of all relevant people represented, even when they express their disagreement with the social worker’s opinion (McDonald et al., 2015; Rai, 2021). This idea coincides with Del Caño et al. (2012), Giribuela and Nieto (2010: 62, 89) and Munuera (2012), among others, who describe the need to incorporate a new variable in the reports: the reference to how the information has been obtained and the sources. An unbiased report should include all points of view and a mention of the source of the opinion. Excluding this information may lead to ignoring the power imbalance between the professional and the person who is likely to be impacted by the recommendations contained in the report itself (Rai, 2021).

This criterion would be valid for any report written in any context or field within social work because it also aims to comply with the precepts of the International Federation of Social Work (IFSW, 2018): social work is a profession that promotes the empowerment of people and their participation in all aspects of the decisions that affect their lives. In short, it is about writing client-centred reports from a well-argued and proven professional point of view.

Despite the efforts to systematise and unify criteria around specific social work instruments (McDonald et al., 2015: 365; Giribuela and Nieto, 2010: 108), more work is needed to ensure that reports stand out for their rigour and professionalism.

Supplemental Material