Overrepresentation of Black, Asian and minority ethnic solicitors in reports to the SRA: the impact of socio-cognitive bias on the likelihood that people will complain about potential misconduct to the SRA

Background to the survey

Our literature review identified potential reasons behind the greater likelihood that Black, Asian and minority ethnic solicitors have complaints about potential misconduct raised about them with the SRA. One such factor identified was the potential that socio-cognitive biases might influence decision-making by some groups of consumers.

This is based on social attribution theory and it is focused on how individuals use information to arrive at causal explanations for events. There are two elements to this theory, which can influence the extent to which an individual may attribute responsibility, which in this context may influence whether or not they make a complaint about someone. These are known as dispositional attribution and situational attribution.

• Dispositional attribution considers the extent to which the perception of misconduct is assigned to an individual's 'internal' characteristics, or a 'deliberate decision' taken by them. Published research suggests this is more likely when a service provider is from a minority ethnic group. Dispositional attribution has been shown in consumer research to increase the likelihood of complaints being made. This may have the effect of amplifying the likelihood of complaints about potential misconduct made to the SRA about Black, Asian and minority ethnic solicitors.
• Situational attribution considers the extent to which the perception of misconduct is due to situations or events outside an individual's control, stressing the importance of 'external' influences or situations 'happening to' the subject. Situational attribution has been shown to reduce the likelihood of complaints being made and to be less likely when a service provider is from a minority ethnic group. This may have the effect of reducing the likelihood of complaints about potential misconduct being made to the SRA about White solicitors.

We wanted to test whether there was any empirical evidence to support this and so we conducted an online survey to understand how consumers attribute responsibility for potential misconduct, when this service is provided by solicitors from different ethnic backgrounds (i.e., Black, Asian, or White solicitors) and gender (i.e., male or female solicitors). We also sought to test whether as a result, these consumers were more likely to report the solicitors in question to the SRA for potential misconduct.

Results by the characteristics of the solicitor

In summary, we found that respondents did not make substantially differential attributions based on either the perceived ethnicity or the gender of the solicitor in the scenario. Nor were they more likely to report a solicitor to the SRA for potential misconduct based on their perceived ethnicity or gender. The results of our analysis by the ethnicity and gender of the solicitors for each question in the survey shows:

• Respondents did not attribute responsibility for the potential misconduct differently depending on the perceived ethnicity or gender of the solicitor in the scenario (see table 6 below).
• When asked whether factors outside the control of the solicitors were responsible (such as bad luck or unforeseeable circumstances), respondents did not respond differently depending on the ethnicity or gender of the solicitor in the scenario (see table 7 below).
• Respondents did not respond differently depending on the ethnicity or gender of the solicitor when asked whether the solicitor who provided the service could have prevented what happened (see table 9 below).
• There was a statistically significant difference by the ethnicity and gender of the solicitor, in responses to whether the solicitor would behave in the same way in the future. In the scenarios involving an Asian-female and White solicitors (both male and female) respondents were more likely to indicate that the solicitors would be very unlikely or unlikely to behave in the same way in the future than for the other scenarios (see table 8 below).
• There was no difference depending on the ethnicity or gender of the solicitor, in whether respondents would report these solicitors to the SRA for potential misconduct (see table 10 below).

Results by the characteristics of the respondents

When we analysed the results by the different characteristics of the respondents, we found evidence that there were differential attributions.

Looking firstly at socio-demographic characteristics we found the following differences:

• The ethnicity of respondents affected how they attributed responsibility for what happened. White respondents ascribed responsibility for what happened to the solicitor in the scenario more than Asian or Black respondents (see table 11 below). There were also some differences by place of birth (see table 16 below), although these characteristics did not affect the likelihood the respondents would report solicitors to the SRA for potential misconduct (see table 15 in relation to ethnicity and 20 in relation to place of birth).
• Female respondents were more likely to attribute responsibility for what happened to the solicitor (see table 21 below), but male respondents were slightly more likely to believe outside factors were mostly responsible (see table 22 below), although there was no significant difference between the genders in terms of the likelihood to report solicitors to the SRA for potential misconduct (see table 25 below).
• Older respondents were more likely to attribute responsibility to the solicitor in the scenario than younger respondents (see table 26 below) and less likely to attribute responsibility to outside factors (see table 27 below). But older respondents were not more or less likely to report them to the SRA for potential misconduct (see table 30 below).
• Respondents not in paid work (e.g., homemaker, retired or disabled) and unemployed respondents were respectively the most and the least likely to attribute responsibility to the solicitor in the scenario (see table 31 below). Employment status also affected the likelihood to report solicitors to the SRA for potential misconduct (see table 35 below).

Results by the respondents' experience of, satisfaction with and knowledge of the legal industry

Looking at the different levels of experience of, satisfaction with, and knowledge of the legal industry that respondents have, we found the following differences:

• Respondents with a higher level of experience of the legal industry were more likely to believe that the potential misconduct could have been prevented than those with a lower level of experience of the legal industry (see table 44 below). This group were also slightly more likely to take the complaint further and report the solicitor to the SRA for potential misconduct (see table 45 below).
• Respondents who were more dissatisfied with the legal industry were more likely to think the solicitor would behave in the same way in the future (in relation to the potential misconduct) than those who were more satisfied with the legal industry (see table 43 below). The level of satisfaction with the legal industry did not significantly affect whether someone would report the solicitor to the SRA for potential misconduct (see table 45 below).
• Respondents with a better legal knowledge were more likely to think that outside factors were totally responsible for what happened than respondents with a poorer legal knowledge. Respondents with a poorer legal knowledge, however, were more likely to think that outside factors were mostly responsible than respondents with a better legal knowledge (see table 47). And respondents with a better legal industry knowledge were more likely to report the solicitor to the SRA for potential misconduct (see table 50 below).

Taken together, the results of our survey provided evidence that respondents attributed responsibility for potential misconduct differently. Different attributions, however, seemed to be linked more to respondents' socio-demographic characteristics and levels of experience of, satisfaction with, and knowledge of the legal industry than to the ethnicity or gender of the solicitor who delivered the service.

To investigate how consumers attribute responsibility for potential misconduct, we administered a survey to a randomised sample of people through Prolific, an online research platform that provides the recruitment and management of participants for online research.

In a pilot survey, administered to 300 people, we tested two slightly different scenarios that we had previously developed drawing on actual complaints received by the SRA. For each scenario, we asked respondents to tell us how likely they would be to submit a complaint to the SRA about potential misconduct if they were in this situation. We then selected the scenario for which the responses were more evenly distributed across the available options, in order to reduce the possibility of different responses being down to factors other than the scenario itself. The scenario we used for the survey is set out in Annex A.

The pilot also tested respondents' assumptions about ethnicity, based on 'typical' names commonly ascribed as belonging to particular ethnic groups. These names were taken from previous research on the impact of ethnicity on customers' complaints (Wood et al., 2009). We tested two names for each ethnicity-gender combination (i.e., two for White-female, two for White-male, two for Black-female, two for Black-male, two for Asian-female, and two for Asian-male). We decided to use these relatively 'high-level' ethnic groups because we wanted to have a large enough sample to analyse. Based on the consistency of the responses we collected in the pilot, we selected one name for each ethnicity-gender combination. The gender of the solicitor in the scenario was indicated by the use of typical pronouns used to indicate whether someone was male or female. The names and pronouns we used to identify the ethnicity and gender of the solicitors in the six scenarios are set out in Annex A.

Following the pilot, we prepared six versions of the same questionnaire, one for each ethnicity-gender combination, with the only differences being the pronoun and name of the solicitor in the scenario. We collected around 700 responses for each of them, resulting in a total sample of around 4,200 responses.

We collected demographic data about the respondents, including ethnicity, place of birth, gender, age and employment status.

Our respondents were almost equally distributed between female (51.5%) and male (48.5%). The largest age group was 18- to 39-years-old (38.8%), followed by 40- to 59-years-old (36.3%), and then 60+ (24.9%). Most of our respondents were White (87.1%) followed by Asian (6.6%) and Black (3.2%) respectively being the second and third largest ethnic group in our sample. Most of our respondents (87.7%) were born in the UK. Respondents born in Europe, Asia, and Africa account for respectively 5.7%, 3.1%, and 2.1% of our sample. Most of our respondents were in full-time employment (41.5%). Respondents not in paid work (e.g., homemaker, retired or disabled) and those in part-time employment account for respectively 17.4% and 15.2% of our sample.

In the first part of the questionnaire, we asked respondents to read the scenario provided and to tell us:

• Q1: To what extent they thought the solicitor in the scenario was responsible for what happened. This question tested the extent to which respondents attribute responsibility for potential misconduct to factors within the control of the solicitor.
• Q2: To what extent they thought factors outside of the control of the solicitor in the scenario (such as bad luck or unforeseeable circumstances) were responsible for what happened. This question tested the extent to which respondents attributed responsibility for potential misconduct to outside factors, such as those mentioned in the question.
• Q3: How likely they thought it was that the solicitor in the scenario would behave in the same way in the future.
• Q4: To what extent they agreed that the solicitor in the scenario could have prevented what happened.
• Q5: How likely they would be to report the solicitor in the scenario to the SRA for potential misconduct.

These questions were asked in order to understand how respondents attributed responsibility and in turn how likely they were to report the solicitor to the SRA for potential misconduct.

In the second part of the questionnaire, we asked respondents several questions, related to their experience of, satisfaction with, and knowledge of the legal industry. More specifically:

• To understand respondents' level of experience of the legal industry, we asked them to answer the following question: 'Have you ever used legal services?' Respondents could choose among the following answers: 'Yes,' 'No,' 'Do not know,' or 'Prefer not to say.' 
• To understand respondents' level of satisfaction with the legal industry, we asked them to answer the following question: 'How satisfied are you with the most recent service you have used?' Respondents were asked to provide an answer on a scale ranging from 1 ('extremely dissatisfied') to 5 ('extremely satisfied').
• To understand respondents' level of knowledge of the legal industry, we asked them to answer the following question: 'To what extent do you agree with the following statement: I have good knowledge and understanding of legal processes?' Respondents were asked to provide an answer on a scale ranging from 1 ('strongly disagree') to 5 ('strongly agree').

These questions were asked in order to understand if these factors affected how respondents attributed responsibility and in turn their likelihood to report the solicitor to the SRA for potential misconduct.

This section looks at the overall response from all respondents across the six scenarios we tested.

Overview of the results

Overall, our respondents thought that the solicitor in the scenario was responsible for what happened, but that outside factors were responsible as well, at least to some extent. Our respondents also thought that the solicitor would be unlikely to behave in the same way in the future and that they could have prevented what happened. Three out of four of our respondents said that they would take the complaint further and report the solicitor to the SRA for potential misconduct.

Detailed results

Table 1 shows the answers our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?'). Most of our respondents thought that the solicitor in the scenario was responsible for what happened, with 43.6% of them saying that the solicitor in the scenario was totally responsible and 30.6% saying that the solicitor was mostly responsible. Only 7.2% of our respondents thought that the solicitor in the scenario was not responsible at all or slightly responsible.

Table 1: Responses to Q1 (responsibility for what happened)

	Frequency	Percent	Valid Percent
Not responsible at all	47	1.1	1.1
Slightly responsible	254	6.1	6.1
Somewhat responsible	775	18.5	18.6
Mostly responsible	1278	30.5	30.6
Totally responsible	1823	43.4	43.6
Total	4177	99.5	100
Missing	19	0.5
Total	4196	100

Table 2 shows the answers our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?'). Most of our respondents thought that outside factors were responsible to some extent for what happened, with 31.3% of them saying that outside factors were slightly responsible and 30.4% saying that outside factors were somewhat responsible. Only 4.9% of our respondents thought that outside factors were totally responsible for what happened.

Table 2: Responses to Q2 (extent to which outside factors were responsible)

	Frequency	Percent	Valid Percent
Outside factors not responsible at all	620	14.8	14.9
Outside factors slightly responsible	1305	31.1	31.3
Outside factors somewhat responsible	1269	30.2	30.4
Outside factors mostly responsible	772	18.4	18.5
Outside factors totally responsible	203	4.8	4.9
Total	4169	99.4	100
Missing	27	0.6
Total	4196	100

Table 3 shows the answers our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?'). Most of our respondents thought that the solicitor in the scenario would be unlikely to behave in the same way in the future, with 41.4% of them saying that the solicitor would be very unlikely to behave in the same way and 33.4% saying that the solicitor would be unlikely to behave in the same way. 11.6% of our respondents, however, believed that the solicitor in the scenario would be likely to behave in the same way, with likely being the third most selected option.

Table 3: Responses to Q3 (likely to behave in the same way in the future)

	Frequency	Percent	Valid Percent
Very unlikely	1700	40.5	41.4
Unlikely	1371	32.7	33.4
Neither likely nor unlikely	337	8	8.2
Likely	476	11.3	11.6
Very likely	225	5.4	5.5
Total	4109	97.9	100
Missing	87	2.1
Total	4196	100

Table 4 shows the answers our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?'). Almost 70% of our respondents strongly agreed that the solicitor in the scenario could have prevented what happened. Only 2% of our respondents either disagreed or strongly disagreed with this statement.

Table 4: Responses to Q4 (extent to which events could have been prevented)

	Frequency	Percent	Valid Percent
Strongly disagree	29	0.7	0.7
Disagree	55	1.3	1.3
Neither agree nor disagree	122	2.9	2.9
Agree	1067	25.4	25.6
Strongly agree	2899	69.1	69.5
Total	4172	99.4	100
Missing	24	0.6
Total	4196	100

Table 5 shows the answers our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?'). Most of our respondents said that they would be either likely (39.0%) or very likely (36.3%) to take the complaint further and report the solicitor in the scenario to the regulator of solicitors. Around 12% of them, however, reported being very unlikely or unlikely to take the complaint further and report the solicitor to the regulator of solicitors.

Table 5: Responses to Q5 (likelihood of reporting to the SRA)

	Frequency	Percent	Valid Percent
Very unlikely	74	1.8	1.8
Unlikely	447	10.7	10.8
Neither likely nor unlikely	500	11.9	12.1
Likely	1608	38.3	39
Very likely	1497	35.7	36.3
Total	4126	98.3	100
Missing	70	1.7
Total	4196	100

This section compares the responses by the gender and ethnicity of the solicitors in the scenario².

Overview of the results

Overall, the answers our respondents provided did not differ across solicitors of different gender and ethnicity in a statistically significant way. Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') was, however, an exception, as the responses provided to this question varied across ethnicity-gender combinations. Some combinations (i.e., Black male, Asian male, Black female) were considered more likely to behave in the same way in the future than others (i.e., Asian female and White female).

Detailed results

Table 6 shows the responses provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') by each individual gender-ethnicity combination in our scenario. The responses provided were rather similar across gender-ethnicity combinations. Most of our respondents thought that the solicitor in the scenario was either totally responsible or mostly responsible for what happened, regardless of gender or ethnicity. Although the analysis found some slight differences, overall, the results of a test we conducted to compare responses across gender-ethnicity combinations were not statistically significant (see Annex B).

Table 6: Responses to Q1 (responsibility for what happened) by gender-ethnicity combination

		Asian female	Asian male	Black female	Black male	White female	White male	Total
Not responsible at all	Count	7	8	10	3	9	10	47
	%	1.00%	1.20%	1.40%	0.40%	1.30%	1.40%	1.10%
Slightly responsible	Count	45	40	36	46	41	46	254
	%	6.50%	5.80%	5.10%	6.60%	5.90%	6.70%	6.10%
Somewhat responsible	Count	119	120	112	151	133	140	775
	%	17.10%	17.30%	16.00%	21.60%	19.10%	20.30%	18.60%
Mostly responsible	Count	223	210	254	205	187	199	1278
	%	32.00%	30.20%	36.20%	29.40%	26.90%	28.80%	30.60%
Totally responsible	Count	302	317	290	293	325	296	1823
	%	43.40%	45.60%	41.30%	42.00%	46.80%	42.80%	43.60%
Total	Count	696	695	702	698	695	691	4177
		100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 7 shows the responses to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') by each gender-ethnicity combination in our scenario. The responses provided are again rather similar across gender-ethnicity combinations, with most of our respondents indicating that outside factors are only slightly or somewhat responsible for what happened. The results of a test we performed to compare responses across gender-ethnicity combinations confirmed that differences in responses are not statistically significant (see Annex B).

Table 7: Responses to Q2 (extent to which outside factors were responsible) by gender-ethnicity combination

		Asian female	Asian male	Black female	Black male	White female	White male	Total
Outside factors not responsible at all	Count	101	117	100	97	106	99	620
	%	14.50%	16.90%	14.30%	14.00%	15.30%	14.30%	14.90%
Outside factors slightly responsible	Count	217	208	231	223	197	229	1305
	%	31.20%	30.00%	33.00%	32.10%	28.40%	33.10%	31.30%
Outside factors somewhat responsible	Count	217	213	197	229	211	202	1269
	%	31.20%	30.70%	28.10%	32.90%	30.40%	29.20%	30.40%
Outside factors mostly responsible	Count	131	122	136	125	135	123	772
	%	18.80%	17.60%	19.40%	18.00%	19.50%	17.80%	18.50%
Outside factors totally responsible	Count	29	33	37	21	45	38	203
	%	4.20%	4.80%	5.30%	3.00%	6.50%	5.50%	4.90%
Total	Count	695	693	701	695	694	691	4169
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 8 shows the responses to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') by each gender-ethnicity combination in our scenario. The responses provided appeared similar across gender-ethnicity combinations in that most of our respondents thought that it was either very unlikely or unlikely that the solicitor in the scenario would behave in the same way in the future, regardless of the gender-ethnicity combination considered. However, those respondents that were presented with a scenario that featured Asian-female, White-female, and White-male solicitors were more likely to indicate that the solicitors in the scenario would be very unlikely or unlikely to behave in the same way in the future than other respondents. The results of a test we performed to compare responses across gender-ethnicity combinations confirmed the relevance of these differences, showing that differences in responses were statistically significant (see Annex B).

Table 8: Responses to Q3 (likely to behave in the same way in the future) by gender-ethnicity combination

		Asian female	Asian male	Black female	Black male	White female	White male	Total
Very unlikely	Count	293	269	268	236	340	294	1700
	%	42.50%	39.40%	39.10%	34.40%	49.70%	43.20%	41.40%
Unlikely	Count	250	217	226	241	213	224	1371
	%	36.20%	31.80%	33.00%	35.10%	31.10%	32.90%	33.40%
Neither likely nor unlikely	Count	48	54	75	64	38	58	337
	%	7.00%	7.90%	10.90%	9.30%	5.60%	8.50%	8.20%
Likely	Count	67	95	79	94	66	75	476
	%	9.70%	13.90%	11.50%	13.70%	9.60%	11.00%	11.60%
Very likely	Count	32	48	37	52	27	29	225
	%	4.60%	7.00%	5.40%	7.60%	3.90%	4.30%	5.50%
Total	Count	690	683	685	687	684	680	4109
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 9 shows the answers our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') by each gender-ethnicity combination in our scenario. The responses provided were quite similar across gender-ethnicity combinations. Most of our respondents either agreed or strongly agreed that the solicitor in the scenario could have prevented what happened, regardless of the gender and ethnicity of the solicitor. The results of a test we performed to compare responses across gender-ethnicity combinations confirmed that differences in responses were not statistically significant (see Annex B).

Table 9: Responses to Q4 (extent to which events could have been prevented) by gender-ethnicity combination

		Asian female	Asian male	Black female	Black male	White female	White male	Total
Strongly disagree	Count	6	5	4	4	4	6	29
	%	0.90%	0.70%	0.60%	0.60%	0.60%	0.90%	0.70%
Disagree	Count	8	8	7	12	7	13	55
	%	1.20%	1.20%	1.00%	1.70%	1.00%	1.90%	1.30%
Neither agree nor disagree	Count	20	14	19	16	24	29	122
	%	2.90%	2.00%	2.70%	2.30%	3.50%	4.20%	2.90%
Agree	Count	193	177	164	190	171	172	1067
	%	27.80%	25.50%	23.40%	27.30%	24.60%	24.90%	25.60%
Strongly agree	Count	468	490	506	475	489	471	2899
	%	67.30%	70.60%	72.30%	68.10%	70.40%	68.20%	69.50%
Total	Count	695	694	700	697	695	691	4172
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 10 shows the answers our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') by each gender-ethnicity combination in our scenario. Most of our respondents reported being either likely or very likely to take the complaint further and report the solicitor in the SRA, regardless of the gender and ethnicity of the solicitor. The results of a test we performed to compare responses across gender-ethnicity combinations confirmed that differences in responses were not statistically significant (see Annex B).

Table 10: Responses to Q5 (likelihood of reporting to the SRA) by gender-ethnicity combination

		Asian female	Asian male	Black female	Black male	White female	White male	Total
Very unlikely	Count	15	11	10	16	10	12	74
	%	2.20%	1.60%	1.40%	2.30%	1.50%	1.70%	1.80%
Unlikely	Count	75	61	80	79	73	79	447
	%	10.90%	9.00%	11.50%	11.50%	10.70%	11.50%	10.80%
Neither likely nor unlikely	Count	89	81	82	87	85	76	500
	%	12.90%	11.90%	11.80%	12.60%	12.40%	11.10%	12.10%
Likely	Count	259	255	276	280	271	267	1608
	%	37.50%	37.40%	39.80%	40.60%	39.60%	38.90%	39.00%
Very likely	Count	253	273	245	227	246	253	1497
	%	36.60%	40.10%	35.40%	32.90%	35.90%	36.80%	36.30%
Total	Count	691	681	693	689	685	687	4126
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

This section looks at the overall response to the scenario by the characteristics of the respondents.

Overview of the results

The responses our respondents provided varied across respondents' profiles. Respondents born in different places or of different ethnicities attributed responsibility for potential misconduct differently. This, however, did not translate into an increased likelihood to submit a complaint to the SRA. In a similar vein, respondents' age and gender affected how they attributed responsibility for potential misconduct, but this again did not translate into a higher likelihood to submit a complaint to the SRA. Respondents of employment status, on the contrary, differed not only in terms of how they attributed responsibility for potential misconduct. They also differed in terms of their likelihood to submit a complaint to the SRA for potential misconduct.

Detailed analysis by respondents' ethnicity

Table 11 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied across their ethnicities. Focusing on the three ethnicities with the largest number of responses (i.e., Asian, Black, and White), White respondents ascribed responsibility to the solicitor in the scenario more for what happened than Asian or Black respondents. White respondents were in fact less likely than Black and Asian respondents to think that the solicitor in the scenario was not responsible at all for what happened and more likely than Black and Asian respondents to think that the solicitor in the scenario was totally responsible for what happened. The results of a test we performed to compare the responses across ethnicities confirmed that the differences observed are statistically significant (see Annex B).

Table 11: Responses to Q1 (responsibility for what happened) by respondent's ethnicity

		Asian	Black	Mixed	Other	White	Total
Not responsible at all	Count	5	4	1	0	37	47
	%	1.80%	3.00%	1.40%	0.00%	1.00%	1.10%
Slightly responsible	Count	22	8	3	6	215	254
	%	7.90%	6.10%	4.20%	10.50%	5.90%	6.10%
Somewhat responsible	Count	64	31	19	13	648	775
	%	23.10%	23.50%	26.80%	22.80%	17.80%	18.60%
Mostly responsible	Count	92	38	16	21	1111	1278
	%	33.20%	28.80%	22.50%	36.80%	30.50%	30.60%
Totally responsible	Count	94	51	32	17	1629	1823
	%	33.90%	38.60%	45.10%	29.80%	44.80%	43.60%
Total	Count	277	132	71	57	3640	4177
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 12 shows that the responses our respondents provided to Q2 (i.e., To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied across their ethnicities. Focusing on the three ethnicities with the largest number of responses (i.e., Asian, Black, and White), Black respondents were more likely than Asian and White respondents to believe that outside factors were somewhat responsible for what happened, but less likely than Asian and White respondents to believe that outside factors were mostly responsible. The results of a test we performed to compare the responses we collected across ethnicities confirm that the differences we observed were statistically significant, although to a lower degree than the responses to Q1 (see Annex B).

Table 12: Responses to Q2 (extent to which outside factors were responsible) by respondent's ethnicity

		Asian	Black	Mixed	Other	White	Total
Outside factors not responsible at all	Count	34	20	11	8	547	620
	%	12.50%	15.30%	15.50%	14.00%	15.00%	14.90%
Outside factors slightly responsible	Count	70	32	23	14	1166	1305
	%	25.60%	24.40%	32.40%	24.60%	32.10%	31.30%
Outside factors somewhat responsible	Count	92	49	23	17	1088	1269
	%	33.70%	37.40%	32.40%	29.80%	29.90%	30.40%
Outside factors mostly responsible	Count	64	20	13	14	661	772
	%	23.40%	15.30%	18.30%	24.60%	18.20%	18.50%
Outside factors totally responsible	Count	13	10	1	4	175	203
	%	4.80%	7.60%	1.40%	7.00%	4.80%	4.90%
Total	Count	273	131	71	57	3637	4169
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 13 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') varied, although only slightly, across their ethnicities. Asian respondents were more likely than Black or White respondents to believe that the solicitor in the scenario would behave in the same way in the future. The results of a test we performed to compare the responses we collected across ethnicities confirmed that the differences we observed were statistically significant, although to a lower degree than for Q1 (see Annex B).

Table 13: Responses to Q3 (likely to behave in the same way in the future) by respondent's ethnicity

		Asian	Black	Mixed	Other	White	Total
Very unlikely	Count	92	53	29	18	1508	1700
	%	33.90%	40.80%	42.00%	32.70%	42.10%	41.40%
Unlikely	Count	95	42	21	20	1193	1371
	%	35.10%	32.30%	30.40%	36.40%	33.30%	33.40%
Neither likely nor unlikely	Count	22	11	10	7	287	337
	%	8.10%	8.50%	14.50%	12.70%	8.00%	8.20%
Likely	Count	44	15	3	9	405	476
	%	16.20%	11.50%	4.30%	16.40%	11.30%	11.60%
Very likely	Count	18	9	6	1	191	225
	%	6.60%	6.90%	8.70%	1.80%	5.30%	5.50%
Total	Count	271	130	69	55	3584	4109
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 14 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied substantially across their ethnicities. White respondents were less likely than Asian or Black respondents to agree with this statement, but more likely than Asian or Black respondents to strongly agree with it. The results of a test we performed to compare the responses we collected across ethnicities confirm that the differences we observed were statistically significant (see Annex B).

Table 14: Responses to Q4 (extent to which events could have been prevented) by respondent's ethnicity

		Asian	Black	Mixed	Other	White	Total
Strongly disagree	Count	3	2	0	1	23	29
	%	1.10%	1.50%	0.00%	1.80%	0.60%	0.70%
Disagree	Count	8	4	0	0	43	55
	%	2.90%	3.00%	0.00%	0.00%	1.20%	1.30%
Neither agree nor disagree	Count	21	4	3	3	91	122
	%	7.60%	3.00%	4.20%	5.30%	2.50%	2.90%
Agree	Count	98	52	21	26	870	1067
	%	35.40%	39.40%	29.60%	45.60%	23.90%	25.60%
Strongly agree	Count	147	70	47	27	2608	2899
	%	53.10%	53.00%	66.20%	47.40%	71.70%	69.50%
Total	Count	277	132	71	57	3635	4172
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 15 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') did not vary substantially across their ethnicities. Although Black respondents were overall more likely than Asian or White respondents to take the complaint further and report the solicitor in the scenario to the regulator of solicitors, the results of a test we performed to compare the responses we collected across ethnicities showed that these differences were not statistically significant (see Annex B).

Table 15: Responses to Q5 (likelihood of reporting to the SRA) by respondent's ethnicity

		Asian	Black	Mixed	Other	White	Total
Very unlikely	Count	5	3	0	2	64	74
	%	1.80%	2.30%	0.00%	3.60%	1.80%	1.80%
Unlikely	Count	23	16	4	4	400	447
	%	8.50%	12.20%	5.60%	7.30%	11.10%	10.80%
Neither likely nor unlikely	Count	39	14	10	13	424	500
	%	14.40%	10.70%	14.10%	23.60%	11.80%	12.10%
Likely	Count	100	45	25	16	1422	1608
	%	36.90%	34.40%	35.20%	29.10%	39.50%	39.00%
Very likely	Count	104	53	32	20	1288	1497
	%	38.40%	40.50%	45.10%	36.40%	35.80%	36.30%
Total	Count	271	131	71	55	3598	4126
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Detailed analysis by respondents' place of birth

Table 16 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied across their places of birth. Focusing on the places accounting for the highest percentages of our respondents (i.e., United Kingdom, Europe, Asia, and Africa), respondents who were born either in the United Kingdom or in Africa were more likely to think that the solicitor in the scenario is totally responsible for what happened than respondents who were born either in Asia or in the rest of Europe. Respondents who were born in Africa were less likely than respondents who were born in the United Kingdom, Asia, or in other European countries to think that the solicitor in the scenario was mainly responsible for what happened, but more likely than respondents born in these other places to think that the solicitor in the scenario was either not responsible at all or only somewhat responsible for what happened. The results of a test we performed to compare the responses we collected across places of birth confirmed that the differences we observed were statistically significant (see Annex B).

Table 16: Responses to Q1 (responsibility for what happened) by respondent's place of birth

		Africa	Asia	Europe	North America	Oceania	South America	United Kingdom	Total
Not responsible at all	Count	4	2	5	1	0	0	35	47
	%	4.50%	1.60%	2.10%	2.60%	0.00%	0.00%	1.00%	1.10%
Slightly responsible	Count	6	12	15	2	0	1	218	254
	%	6.80%	9.40%	6.30%	5.10%	0.00%	14.30%	5.90%	6.10%
Somewhat responsible	Count	23	22	55	9	1	1	664	775
	%	26.10%	17.20%	23.10%	23.10%	7.70%	14.30%	18.10%	18.60%
Mostly responsible	Count	18	45	75	14	5	4	1117	1278
	%	20.50%	35.20%	31.50%	35.90%	38.50%	57.10%	30.50%	30.60%
Totally responsible	Count	37	47	88	13	7	1	1630	1823
	%	42.00%	36.70%	37.00%	33.30%	53.80%	14.30%	44.50%	43.60%
Total	Count	88	128	238	39	13	7	3664	4177
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 17 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied significantly across their places of birth. Focusing on the places accounting for the highest percentages of our respondents (i.e., United Kingdom, Europe, Asia, and Africa), respondents who were born in Asia were more likely to think that factors outside of the control of the solicitor in the scenario were either not responsible at all or only slightly responsible for what happened than respondents who were born in Africa, the United Kingdom, or other European countries. Respondents who were born in Africa were more likely than respondents who were born in Asia, the United Kingdom, or other European countries to think that factors outside of the control of the solicitor in the scenario were totally responsible for what happened. The results of a test we performed to compare the responses we collected across places of birth confirmed that the differences we observed were statistically significant (see Annex B).

Table 17: Responses to Q2 (extent to which outside factors were responsible) by respondent's place of birth

		Africa	Asia	Europe	North America	Oceania	South America	United Kingdom	Total
Outside factors not responsible at all	Count	15	23	26	4	2	0	550	620
	%	17.20%	18.30%	10.90%	10.30%	15.40%	0.00%	15.00%	14.90%
Outside factors slightly responsible	Count	16	38	56	14	5	3	1173	1305
	%	18.40%	30.20%	23.50%	35.90%	38.50%	42.90%	32.10%	31.30%
Outside factors somewhat responsible	Count	28	30	72	10	4	3	1122	1269
	%	32.20%	23.80%	30.30%	25.60%	30.80%	42.90%	30.70%	30.40%
Outside factors mostly responsible	Count	15	28	61	8	0	1	659	772
	%	17.20%	22.20%	25.60%	20.50%	0.00%	14.30%	18.00%	18.50%
Outside factors totally responsible	Count	13	7	23	3	2	0	155	203
	%	14.90%	5.60%	9.70%	7.70%	15.40%	0.00%	4.20%	4.90%
Total	Count	87	126	238	39	13	7	3659	4169
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 18 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') varied considerably across their places of birth. Focusing on the places accounting for the highest percentages of our respondents (i.e., United Kingdom, Europe, Asia, and Africa), respondents who were born in Asia were more likely than respondents born in Africa, the United Kingdom, or other European countries to think that the solicitor in the scenario would behave in the same way in the future. The results of a test we performed to compare the responses we collected across places of birth confirmed that the differences we observed were statistically significant (see Annex B).

Table 18: Responses to Q3 (likely to behave in the same way in the future) by respondent's place of birth

		Africa	Asia	Europe	North America	Oceania	South America	United Kingdom	Total
Very unlikely	Count	37	26	100	13	3	4	1517	1700
	%	42.50%	20.60%	42.00%	35.10%	25.00%	57.10%	42.10%	41.40%
Unlikely	Count	24	45	71	13	6	1	1211	1371
	%	27.60%	35.70%	29.80%	35.10%	50.00%	14.30%	33.60%	33.40%
Neither likely nor unlikely	Count	9	21	28	2	1	2	274	337
	%	10.30%	16.70%	11.80%	5.40%	8.30%	28.60%	7.60%	8.20%
Likely	Count	12	23	25	5	2	0	409	476
	%	13.80%	18.30%	10.50%	13.50%	16.70%	0.00%	11.40%	11.60%
Very likely	Count	5	11	14	4	0	0	191	225
	%	5.70%	8.70%	5.90%	10.80%	0.00%	0.00%	5.30%	5.50%
Total	Count	87	126	238	37	12	7	3602	4109
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 19 shows that the responses our respondents provide to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied significantly across their places of birth. Focusing on the places accounting for the highest percentages of our respondents (i.e., United Kingdom, Europe, Asia, and Africa), respondents who were born in Africa were more likely to either strongly disagree or disagree with this statement than respondents who were born in Asia, the United Kingdom, or other European countries. On the contrary, respondents who were born in the United Kingdom totally agreed more with this statement than respondents who were born in Asia, Africa, or other European countries. The results of a test we performed to compare the responses we collected across places of birth confirmed that these differences were statistically significant (see Annex B).

Table 19: Responses to Q4 (extent to which events could have been prevented) by respondent's place of birth

		Africa	Asia	Europe	North America	Oceania	South America	United Kingdom	Total
Strongly disagree	Count	3	2	1	1	0	0	22	29
	%	3.40%	1.60%	0.40%	2.60%	0.00%	0.00%	0.60%	0.70%
Disagree	Count	3	3	8	0	0	0	41	55
	%	3.40%	2.30%	3.40%	0.00%	0.00%	0.00%	1.10%	1.30%
Neither agree nor disagree	Count	3	7	12	0	0	0	100	122
	%	3.40%	5.50%	5.00%	0.00%	0.00%	0.00%	2.70%	2.90%
Agree	Count	36	46	66	12	1	4	902	1067
	%	40.90%	35.90%	27.70%	30.80%	7.70%	57.10%	24.70%	25.60%
Strongly agree	Count	43	70	151	26	12	3	2594	2899
	%	48.90%	54.70%	63.40%	66.70%	92.30%	42.90%	70.90%	69.50%
Total	Count	88	128	238	39	13	7	3659	4172
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 20 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') did not vary substantially across their places of birth. Focusing on the countries accounting for the highest percentages of our respondents (i.e., United Kingdom, Europe, Asia, and Africa), respondents who were born in Asia were slightly more likely to complain than respondents who were born in Africa, the United Kingdom, or other European countries. The results of a test we performed to compare the responses we collected across places of birth, however, showed that the differences we observed were not statistically significant (see Annex B).

Table 20: Responses to Q5 (likelihood of reporting to the SRA) by respondent's place of birth

		Africa	Asia	Europe	North America	Oceania	South America	United Kingdom	Total
Very unlikely	Count	4	2	2	0	0	0	66	74
	%	4.60%	1.60%	0.90%	0.00%	0.00%	0.00%	1.80%	1.80%
Unlikely	Count	9	7	35	5	1	1	389	447
	%	10.30%	5.60%	15.00%	13.50%	7.70%	14.30%	10.70%	10.80%
Neither likely nor unlikely	Count	14	21	30	5	1	2	427	500
	%	16.10%	16.90%	12.90%	13.50%	7.70%	28.60%	11.80%	12.10%
Likely	Count	26	42	81	17	4	2	1436	1608
	%	29.90%	33.90%	34.80%	45.90%	30.80%	28.60%	39.60%	39.00%
Very likely	Count	34	52	85	10	7	2	1307	1497
	%	39.10%	41.90%	36.50%	27.00%	53.80%	28.60%	36.10%	36.30%
Total	Count	87	124	233	37	13	7	3625	4126
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Detailed analysis by respondents' gender

Table 21 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied across genders. Female respondents were in fact more likely to think that the solicitor in the scenario was responsible for what happened than male respondents. The results of a test we performed to compare the responses we collected across genders confirmed that the differences we observed were statistically significant (see Annex B).

Table 21: Responses to Q1 (responsibility for what happened) by respondent's gender

		Female	Male	Total
Not responsible at all	Count	17	30	47
	%	0.80%	1.50%	1.10%
Slightly responsible	Count	116	138	254
	%	5.40%	6.80%	6.10%
Somewhat responsible	Count	367	408	775
	%	17.10%	20.10%	18.60%
Mostly responsible	Count	673	605	1278
	%	31.30%	29.90%	30.60%
Totally responsible	Count	978	845	1823
	%	45.50%	41.70%	43.60%
Total	Count	2151	2026	4177
	%	100.00%	100.00%	100.00%

Table 22 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') were rather similar across genders, although male respondents were slightly more likely to believe that outside factors were mostly responsible for what happened than female respondents. The results of a test we performed to compare the responses we collected across genders, however, confirmed that the differences we observed were not statistically significant (see Annex B).

Table 22: Responses to Q2 (extent to which outside factors were responsible) by respondent's gender

		Female	Male	Total
Outside factors not responsible at all	Count	319	301	620
	%	14.90%	14.90%	14.90%
Outside factors slightly responsible	Count	682	623	1305
	%	31.80%	30.80%	31.30%
Outside factors somewhat responsible	Count	664	605	1269
	%	30.90%	29.90%	30.40%
Outside factors mostly responsible	Count	382	390	772
	%	17.80%	19.30%	18.50%
Outside factors totally responsible	Count	101	102	203
	%	4.70%	5.00%	4.90%
Total	Count	2148	2021	4169
	%	100.00%	100.00%	100.00%

Table 23 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') were quite similar across genders, but that female respondents were less likely to believe that the solicitor in the scenario would behave in the same way in the future than male respondents. The results of a test we performed to compare responses across genders confirmed that the differences we observed were statistically significant, although only marginally (see Annex B).

Table 23: Responses to Q3 (likely to behave in the same way in future) by respondent's gender

		Female	Male	Total
Very unlikely	Count	905	795	1700
	%	42.70%	39.90%	41.40%
Unlikely	Count	698	673	1371
	%	32.90%	33.80%	33.40%
Neither likely nor unlikely	Count	172	165	337
	%	8.10%	8.30%	8.20%
Likely	Count	234	242	476
	%	11.00%	12.20%	11.60%
Very likely	Count	110	115	225
	%	5.20%	5.80%	5.50%
Total	Count	2119	1990	4109
	%	100.00%	100.00%	100.00%

Table 24 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') were quite similar across genders. The results of a test we performed to compare the responses across genders confirmed that the differences we observed were not statistically significant (see Annex B).

Table 24: Responses to Q4 (extent to which events could have been prevented) by respondent's gender

		Female	Male	Total
Strongly disagree	Count	19	10	29
	%	0.90%	0.50%	0.70%
Disagree	Count	27	28	55
	%	1.30%	1.40%	1.30%
Neither agree nor disagree	Count	53	69	122
	%	2.50%	3.40%	2.90%
Agree	Count	539	528	1067
	%	25.10%	26.10%	25.60%
Strongly agree	Count	1511	1388	2899
	%	70.30%	68.60%	69.50%
Total	Count	2149	2023	4172
	%	100.00%	100.00%	100.00%

Table 25 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') were quite similar across genders. The results of a test we performed to compare the responses across genders confirmed that the differences we observed were not statistically significant (see Annex B).

Table 25: Responses to Q5 (likelihood of reporting to the SRA) by respondent's gender

		Female	Male	Total
Very unlikely	Count	30	44	74
	%	1.40%	2.20%	1.80%
Unlikely	Count	219	228	447
	%	10.30%	11.40%	10.80%
Neither likely nor unlikely	Count	245	255	500
	%	11.60%	12.70%	12.10%
Likely	Count	862	746	1608
	%	40.70%	37.20%	39.00%
Very likely	Count	764	733	1497
	%	36.00%	36.50%	36.30%
Total	Count	2120	2006	4126
	%	100.00%	100.00%	100.00%

Detailed analysis by respondents' age

Table 26 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied quite considerably across their ages. Younger respondents were more likely to think that the solicitor in the scenario was either not responsible at all or only slightly responsible for what happened, whereas older respondents were more likely to think that the solicitor in the scenario is totally responsible for what happened. The results of a test we performed to compare the responses we collected across ages confirmed that these differences were statistically significant (see Annex B).

Table 26: Responses to Q1 (responsibility for what happened) by respondent's age

		18-39	40-59	60+	Total
Not responsible at all	Count	28	15	4	47
	%	1.70%	1.00%	0.40%	1.10%
Slightly responsible	Count	150	75	29	254
	%	9.30%	4.90%	2.80%	6.10%
Somewhat responsible	Count	419	242	114	775
	%	25.80%	15.90%	11.00%	18.60%
Mostly responsible	Count	551	446	281	1278
	%	34.00%	29.40%	27.10%	30.60%
Totally responsible	Count	473	740	610	1823
	%	29.20%	48.70%	58.80%	43.60%
Total	Count	1621	1518	1038	4177
	%	100.00%	100.00%	100.00%	100.00%

Table 27 shows that the responses our respondents provided to Q2 ('To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied considerably across their ages. Younger respondents were more likely to think that outside factors were either somehow or mostly responsible for what happened, whereas older respondents were more likely to think that outside factors were either not responsible or slightly responsible for what happened. The results of a test we performed to compare the responses we collected across ages confirmed that these differences were statistically significant (see Annex B).

Table 27: Responses to Q2 (extent to which outside factors were responsible) by respondent's age

		18-39	40-59	60+	Total
Outside factors not responsible at all	Count	123	248	249	620
	%	7.60%	16.40%	24.00%	14.90%
Outside factors slightly responsible	Count	455	486	364	1305
	%	28.10%	32.10%	35.10%	31.30%
Outside factors somewhat responsible	Count	569	447	253	1269
	%	35.10%	29.50%	24.40%	30.40%
Outside factors mostly responsible	Count	405	245	122	772
	%	25.00%	16.20%	11.80%	18.50%
Outside factors totally responsible	Count	67	87	49	203
	%	4.10%	5.80%	4.70%	4.90%
Total	Count	1619	1513	1037	4169
	%	100.00%	100.00%	100.00%	100.00%

Table 28 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') did not substantially vary across their ages. Although older respondents were more likely to think that the solicitor in the scenario was very unlikely to behave in the same way in the future than younger respondents, the results of a test we performed to compare the responses we collected across ages showed that the differences we observed were not statistically significant (see Annex B).

Table 28: Responses to Q3 (likely to behave in the same way in the future) by respondent's age

		18-39	40-59	60+	Total
Very unlikely	Count	643	609	448	1700
	%	40.20%	40.80%	44.00%	41.40%
Unlikely	Count	547	515	309	1371
	%	34.20%	34.50%	30.30%	33.40%
Neither likely nor unlikely	Count	134	115	88	337
	%	8.40%	7.70%	8.60%	8.20%
Likely	Count	194	169	113	476
	%	12.10%	11.30%	11.10%	11.60%
Very likely	Count	80	84	61	225
	%	5.00%	5.60%	6.00%	5.50%
Total	Count	1598	1492	1019	4109
	%	100.00%	100.00%	100.00%	100.00%

Table 29 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied across their ages. In particular, younger respondents were more likely to agree with the statement in this question, while older respondents were more likely to strongly agree with this statement. The results of a test we performed to compare the responses we collected across ages confirmed that these differences were statistically significant (see Annex B).

Table 29: Responses to Q4 (extent to which events could have been prevented) by respondent's age

		18-39	40-59	60+	Total
Strongly disagree	Count	10	10	9	29
	%	0.60%	0.70%	0.90%	0.70%
Disagree	Count	41	11	3	55
	%	2.50%	0.70%	0.30%	1.30%
Neither agree nor disagree	Count	74	34	14	122
	%	4.60%	2.20%	1.40%	2.90%
Agree	Count	554	361	152	1067
	%	34.20%	23.80%	14.70%	25.60%
Strongly agree	Count	940	1100	859	2899
	%	58.10%	72.60%	82.80%	69.50%
Total	Count	1619	1516	1037	4172
	%	100.00%	100.00%	100.00%	100.00%

Table 30 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') did not substantially vary across their ages, with most of our respondents being either likely or very likely to take the complaint further and report the solicitor in the scenario to the regulator of solicitors. The results of a test we performed to compare the responses we collected across ages confirmed that the differences we observed were not statistically significant (see Annex B).

Table 30: Responses to Q5 (likelihood of reporting to the SRA) by respondent's age

		18-39	40-59	60+	Total
Very unlikely	Count	28	21	25	74
	%	1.70%	1.40%	2.40%	1.80%
Unlikely	Count	166	185	96	447
	%	10.30%	12.40%	9.40%	10.80%
Neither likely nor unlikely	Count	173	186	141	500
	%	10.80%	12.40%	13.80%	12.10%
Likely	Count	659	541	408	1608
	%	41.00%	36.20%	39.90%	39.00%
Very likely	Count	581	563	353	1497
	%	36.20%	37.60%	34.50%	36.30%
Total	Count	1607	1496	1023	4126
	%	100.00%	100.00%	100.00%	100.00%

Detailed analysis by respondents' employment status

Table 31 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied quite considerably across their employment statuses. While respondents in full- or part-time jobs were more likely to think that the solicitor in the scenario is totally responsible for what happened, respondents who are about to start a new job or unemployed were more likely to think that the solicitor in the scenario was mostly responsible for what happened. The results of a test we performed to compare the responses we collected across employment statuses confirmed that the differences we observed were statistically significant (see Annex B).

Table 31: Responses to Q1 (responsibility for what happened) by respondent's employment status

		Not in paid work	Unemployed	Part-time	Full-time	Other	Due to start a new job	Total
Not responsible at all	Count	5	6	10	18	1	0	40
	%	0.70%	4.30%	1.60%	1.00%	0.90%	0.00%	1.20%
Slightly responsible	Count	31	11	35	116	8	1	202
	%	4.30%	7.80%	5.50%	6.70%	7.30%	4.50%	6.00%
Somewhat responsible	Count	87	41	111	333	26	6	604
	%	12.00%	29.10%	17.50%	19.20%	23.60%	27.30%	17.90%
Mostly responsible	Count	212	50	180	535	41	9	1027
	%	29.20%	35.50%	28.40%	30.80%	37.30%	40.90%	30.50%
Totally responsible	Count	390	33	297	733	34	6	1493
	%	53.80%	23.40%	46.90%	42.20%	30.90%	27.30%	44.40%
Total	Count	725	141	633	1735	110	22	3366
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 32 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied quite considerably across their employment statuses. While respondents in full- or part-time jobs were more likely to think that factors outside of the control of the solicitor in the scenario were either somewhat or mostly responsible for what happened, respondents who were about to start a new job or unemployed were more likely to think that factors outside of the control of the solicitor in the scenario were totally responsible for what happened. The results of a test we performed to compare the responses we collected across employment statuses confirmed that the differences we observed were statistically significant (see Annex B).

Table 32: Responses to Q2 (extent to which outside factors were responsible) by respondent's employment status

		Not in paid work	Unemployed	Part-time	Full-time	Other	Due to start a new job	Total
Outside factors not responsible at all	Count	123	10	105	258	14	1	511
	%	17.00%	7.10%	16.60%	14.90%	12.80%	4.50%	15.20%
Outside factors slightly responsible	Count	267	37	193	530	23	7	1057
	%	36.80%	26.20%	30.50%	30.60%	21.10%	31.80%	31.40%
Outside factors somewhat responsible	Count	203	41	188	532	47	5	1016
	%	28.00%	29.10%	29.70%	30.70%	43.10%	22.70%	30.20%
Outside factors mostly responsible	Count	98	45	107	339	16	8	613
	%	13.50%	31.90%	16.90%	19.60%	14.70%	36.40%	18.20%
Outside factors totally responsible	Count	34	8	40	72	9	1	164
	%	4.70%	5.70%	6.30%	4.20%	8.30%	4.50%	4.90%
Total	Count	725	141	633	1731	109	22	3361
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 33 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') did not vary substantially across their employment statuses. Respondents who were about to start a new job or unemployed were slightly more likely to think that the solicitor in the scenario would not behave in the same way in the future, while full-time employees were slightly more likely to think that the solicitor in the scenario would behave in the same way in the future. The results of a test we performed to compare the responses we collected across employment statuses, however, showed that these differences were not statistically significant (see Annex B).

Table 33: Responses to Q3 (likely to behave in the same way in the future) by respondent's employment status

		Not in paid work	Unemployed	Part-time	Full-time	Other	Due to start a new job	Total
Very unlikely	Count	309	52	266	701	44	11	1383
	%	43.50%	38.80%	42.60%	41.00%	40.00%	50.00%	41.80%
Unlikely	Count	231	51	194	582	38	8	1104
	%	32.50%	38.10%	31.00%	34.10%	34.50%	36.40%	33.40%
Neither likely nor unlikely	Count	55	12	57	116	8	0	248
	%	7.70%	9.00%	9.10%	6.80%	7.30%	0.00%	7.50%
Likely	Count	76	15	68	225	14	2	400
	%	10.70%	11.20%	10.90%	13.20%	12.70%	9.10%	12.10%
Very likely	Count	39	4	40	85	6	1	175
	%	5.50%	3.00%	6.40%	5.00%	5.50%	4.50%	5.30%
Total	Count	710	134	625	1709	110	22	3310
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 34 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied quite considerably across their employment statuses. Respondents who were about to start a new job were the least likely to agree with the statement in the question, while respondents not in paid work (e.g., homemakers, retired or disabled) were the most likely to agree with this statement. The results of a test we performed to compare the responses we collected across employment statuses confirmed that the differences we observed were statistically significant (see Annex B).

Table 34: Responses to Q4 (extent to which events could have been prevented) by respondent's employment status

		Not in paid work	Unemployed	Part-time	Full-time	Other	Due to start a new job	Total
Strongly disagree	Count	7	0	3	12	0	0	22
	%	1.00%	0.00%	0.50%	0.70%	0.00%	0.00%	0.70%
Disagree	Count	5	7	16	16	4	0	48
	%	0.70%	5.00%	2.50%	0.90%	3.60%	0.00%	1.40%
Neither agree nor disagree	Count	20	7	14	54	2	2	99
	%	2.80%	5.00%	2.20%	3.10%	1.80%	9.10%	2.90%
Agree	Count	123	55	158	451	35	12	834
	%	17.00%	39.00%	25.00%	26.00%	31.80%	54.50%	24.80%
Strongly agree	Count	568	72	442	1201	69	8	2360
	%	78.60%	51.10%	69.80%	69.30%	62.70%	36.40%	70.20%
Total	Count	723	141	633	1734	110	22	3363
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 35 show that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?) varied slightly across their employment statuses. Respondents who were about to start a new job or were unemployed were the least likely to take the complaint further and report the solicitor in the scenario to the regulator of solicitors. The results of a test we performed to compare the responses we collected across employment statuses confirmed that the differences we observed were statistically significant (see Annex B).

Table 35: Responses to Q5 (likelihood of reporting to the SRA) by respondent's employment status

		Not in paid work	Unemployed	Part-time	Full-time	Other	Due to start a new job	Total
Very unlikely	Count	14	5	10	30	2	0	61
	%	2.00%	3.50%	1.60%	1.70%	1.80%	0.00%	1.80%
Unlikely	Count	74	19	73	168	14	4	352
	%	10.40%	13.50%	11.70%	9.80%	12.80%	18.20%	10.60%
Neither likely nor unlikely	Count	100	17	75	186	10	2	390
	%	14.00%	12.10%	12.00%	10.80%	9.20%	9.10%	11.70%
Likely	Count	290	59	232	653	50	11	1295
	%	40.60%	41.80%	37.20%	38.10%	45.90%	50.00%	39.00%
Very likely	Count	236	41	233	678	33	5	1226
	%	33.10%	29.10%	37.40%	39.50%	30.30%	22.70%	36.90%
Total	Count	714	141	623	1715	109	22	3324
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Overview of the results

The responses provided varied across levels of experience of, satisfaction with, and knowledge of the legal industry.

Respondents with different levels of satisfaction with the legal industry attributed responsibility for potential misconduct differently, but this was not reflected in an increased likelihood to report the potential misconduct to the SRA.

Respondents with different levels of experience and knowledge of the legal industry, on the contrary, not only attributed responsibility for potential misconduct differently. They also varied in their likelihood to report this potential misconduct to the SRA.

Detailed analysis by respondents' level of experience of the legal industry

Table 36 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied quite considerably across their levels of experience of the legal industry. Respondents with a higher level of experience were more likely to think that the solicitor in the scenario was totally responsible for what happened. The results of a test we performed to compare the responses we collected across our respondents' levels of experience of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 36: Responses to Q1 (responsibility for what happened) by respondent's level of experience of the legal industry

		Yes	No	Prefer not to say	Total
Not responsible at all	Count	23	23	1	47
	%	0.80%	1.70%	2.00%	1.10%
Slightly responsible	Count	130	121	1	252
	%	4.70%	9.00%	2.00%	6.10%
Somewhat responsible	Count	408	338	15	761
	%	14.80%	25.20%	30.00%	18.40%
Mostly responsible	Count	798	446	18	1262
	%	29.00%	33.30%	36.00%	30.50%
Totally responsible	Count	1389	411	15	1815
	%	50.50%	30.70%	30.00%	43.90%
Total	Count	2748	1339	50	4137
	%	100.00%	100.00%	100.00%	100.00%

Table 37 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied again quite considerably across their levels of experience of the legal industry. Respondents with a higher level of experience were more likely to think that factors outside of the control of the solicitor in the scenario were either somewhat or mostly responsible for what happened, while respondents with a lower level of experience were more likely to think that factors outside of the control of the solicitor in the scenario were either not responsible or only slightly responsible for what happened. The results of a test we performed to compare the responses we collected across our respondents' levels of experience of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 37: Responses to Q2 (extent to which outside factors were responsible) by respondent's level of experience of the legal industry

		Yes	No	Prefer not to say	Total
Outside factors not responsible at all	Count	490	121	5	616
	%	17.90%	9.10%	10.00%	14.90%
Outside factors slightly responsible	Count	891	396	12	1299
	%	32.50%	29.60%	24.00%	31.50%
Outside factors somewhat responsible	Count	809	428	19	1256
	%	29.50%	32.00%	38.00%	30.40%
Outside factors mostly responsible	Count	411	336	12	759
	%	15.00%	25.10%	24.00%	18.40%
Outside factors totally responsible	Count	142	55	2	199
	%	5.20%	4.10%	4.00%	4.80%
Total	Count	2743	1336	50	4129
	%	100.00%	100.00%	100.00%	100.00%

Table 38 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') did not vary substantially across their levels of experience of the legal industry. The results of a test we performed to compare the responses we collected across our respondents' levels of experience of the legal industry confirmed that the differences we observed were not statistically significant (see Annex B).

Table 38: Responses to Q3 (likely to behave in the same way in the future) by respondent's level of experience of the legal industry

		Yes	No	Prefer not to say	Total
Very unlikely	Count	1102	559	21	1682
	%	40.80%	42.40%	42.00%	41.30%
Unlikely	Count	892	448	17	1357
	%	33.00%	34.00%	34.00%	33.30%
Neither likely nor unlikely	Count	232	100	2	334
	%	8.60%	7.60%	4.00%	8.20%
Likely	Count	322	148	3	473
	%	11.90%	11.20%	6.00%	11.60%
Very likely	Count	153	64	7	224
	%	5.70%	4.90%	14.00%	5.50%
Total	Count	2701	1319	50	4070
	%	100.00%	100.00%	100.00%	100.00%

Table 39 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied quite substantially across their levels of experience of the legal industry. In particular, respondents with a higher level of experience were more likely to totally agree that the solicitor in the scenario could have prevented what happened, while respondents with a lower level of experience were more likely to either disagree or strongly disagree that the solicitor in the scenario could have prevented what happened. The results of a test we performed to compare the responses we collected across our respondents' levels of experience of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 39: Responses to Q4 (extent to which events could have been prevented) by respondent's level of experience of the legal industry

		Yes	No	Prefer not to say	Total
Strongly disagree	Count	17	11	1	29
	%	0.60%	0.80%	2.00%	0.70%
Disagree	Count	26	28	0	54
	%	0.90%	2.10%	0.00%	1.30%
Neither agree nor disagree	Count	65	50	4	119
	%	2.40%	3.70%	8.00%	2.90%
Agree	Count	596	433	17	1046
	%	21.70%	32.30%	34.00%	25.30%
Strongly agree	Count	2039	817	28	2884
	%	74.30%	61.00%	56.00%	69.80%
Total	Count	2743	1339	50	4132
	%	100.00%	100.00%	100.00%	100.00%

Table 40 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') varied, although quite slightly, across our respondents' levels of experience of the legal industry. Respondents with a higher level of experience were slightly more likely to take the complaint further and report the solicitor to the regulator of solicitors than respondents with a lower level of experience. The results of a test we performed to compare the responses we collected across our respondents' levels of experience of the legal industry confirmed that the differences we observed were statistically significant, although to a lesser extent than observed for Q1, Q2, and Q4 (see Annex B).

Table 40: Responses to Q5 (likelihood of reporting to the SRA) by respondent's level of experience of the legal industry

		Yes	No	Prefer not to say	Total
Very unlikely	Count	47	23	3	73
	%	1.70%	1.70%	6.30%	1.80%
Unlikely	Count	284	155	3	442
	%	10.50%	11.70%	6.30%	10.80%
Neither likely nor unlikely	Count	316	169	6	491
	%	11.60%	12.80%	12.50%	12.00%
Likely	Count	1031	547	19	1597
	%	38.00%	41.30%	39.60%	39.10%
Very likely	Count	1038	430	17	1485
	%	38.20%	32.50%	35.40%	36.30%
Total	Count	2716	1324	48	4088
	%	100.00%	100.00%	100.00%	100.00%

Detailed analysis by respondent's level of satisfaction with the legal industry

Table 41 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied quite considerably across their levels of satisfaction with the legal industry. Respondents who were very satisfied were the least likely to think that the solicitor in the scenario was totally responsible for what happened and the most likely to think that the solicitor in the scenario was either not responsible at all or only slightly responsible for what happened. The results of a test we performed to compare the responses we collected across our respondents' levels of satisfaction with the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 41: Responses to Q1 (responsibility for what happened) by respondent's level of satisfaction with the legal industry

		Very satisfied	Satisfied	Neither satisfied nor dissatisfied	Dissatisfied	Very dissatisfied	Total
Not responsible at all	Count	0	3	3	12	5	23
	%	0.00%	1.60%	0.90%	1.00%	0.50%	0.80%
Slightly responsible	Count	2	9	17	62	39	129
	%	3.20%	4.90%	5.20%	5.10%	4.10%	4.70%
Somewhat responsible	Count	11	31	57	196	113	408
	%	17.50%	16.80%	17.40%	16.10%	12.00%	14.90%
Mostly responsible	Count	16	50	86	383	261	796
	%	25.40%	27.20%	26.20%	31.40%	27.60%	29.10%
Totally responsible	Count	34	91	165	568	526	1384
	%	54.00%	49.50%	50.30%	46.50%	55.70%	50.50%
Total	Count	63	184	328	1221	944	2740
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 42 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') did not vary substantially across their levels of satisfaction with the legal industry. The results of a test we performed to compare the responses we collected across our respondents' levels of satisfaction with the legal industry confirmed that the differences we observed were not statistically significant (see Annex B).

Table 42: Responses to Q2 (extent to which outside factors were responsible) by respondent's level of satisfaction with the legal industry

		Very satisfied	Satisfied	Neither satisfied nor dissatisfied	Dissatisfied	Very dissatisfied	Total
Outside factors not responsible at all	Count	17	30	57	198	186	488
	%	27.00%	16.30%	17.50%	16.20%	19.70%	17.80%
Outside factors slightly responsible	Count	9	63	106	395	312	885
	%	14.30%	34.20%	32.50%	32.40%	33.10%	32.40%
Outside factors somewhat responsible	Count	21	50	101	389	248	809
	%	33.30%	27.20%	31.00%	31.90%	26.30%	29.60%
Outside factors mostly responsible	Count	9	30	46	187	139	411
	%	14.30%	16.30%	14.10%	15.30%	14.80%	15.00%
Outside factors totally responsible	Count	7	11	16	51	57	142
	%	11.10%	6.00%	4.90%	4.20%	6.10%	5.20%
Total	Count	63	184	326	1220	942	2735
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 43 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') varied across their levels of satisfaction with the legal industry. In particular, the more dissatisfied our respondents were, the more likely they were to think that the solicitor in the scenario would be unlikely – rather than very unlikely – to behave in the same way in the future. The results of a test we performed to compare the responses we collected across our respondents' levels of satisfaction with the legal industry confirmed that the differences we observed are statistically significant, although to a lesser extent than observed for Q1 (see Annex B).

Table 43: Responses to Q3 (likely to behave in the same way in the future) by respondent's level of satisfaction with the legal industry

		Very satisfied	Satisfied	Neither satisfied nor dissatisfied	Dissatisfied	Very dissatisfied	Total
Very unlikely	Count	27	75	107	467	424	1100
	%	45.80%	41.20%	33.20%	38.80%	45.70%	40.80%
Unlikely	Count	15	51	115	435	274	890
	%	25.40%	28.00%	35.70%	36.10%	29.50%	33.00%
Neither likely nor unlikely	Count	4	20	32	109	66	231
	%	6.80%	11.00%	9.90%	9.10%	7.10%	8.60%
Likely	Count	4	28	41	141	107	321
	%	6.80%	15.40%	12.70%	11.70%	11.50%	11.90%
Very likely	Count	9	8	27	52	57	153
	%	15.30%	4.40%	8.40%	4.30%	6.10%	5.70%
Total	Count	59	182	322	1204	928	2695
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 44 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied rather substantially across their levels of satisfaction with the legal industry. In particular, respondents who were very dissatisfied strongly agreed more with the statement in the question than other respondents. The results of a test we performed to compare the responses we collected across our respondents' levels of satisfaction with the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 44: Responses to Q4 (extent to which events could have been prevented) by respondent's level of satisfaction with the legal industry

		Very satisfied	Satisfied	Neither satisfied nor dissatisfied	Dissatisfied	Very dissatisfied	Total
Strongly disagree	Count	0	1	1	7	8	17
	%	0.00%	0.50%	0.30%	0.60%	0.90%	0.60%
Disagree	Count	0	4	3	12	7	26
	%	0.00%	2.20%	0.90%	1.00%	0.70%	1.00%
Neither agree nor disagree	Count	2	3	14	34	12	65
	%	3.20%	1.60%	4.30%	2.80%	1.30%	2.40%
Agree	Count	14	41	72	310	157	594
	%	22.20%	22.30%	22.00%	25.40%	16.70%	21.70%
Strongly agree	Count	47	135	238	857	756	2033
	%	74.60%	73.40%	72.60%	70.20%	80.40%	74.30%
Total	Count	63	184	328	1220	940	2735
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 45 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') were rather similar across our respondents' levels of satisfaction with the legal industry. The results of a test we performed to compare the responses we collected across our respondents' levels of satisfaction with the legal industry confirmed that the differences we observed were not statistically significant (see Annex B).

Table 45: Responses to Q5 (likelihood of reporting to the SRA) by respondent's level of satisfaction with the legal industry

		Very satisfied	Satisfied	Neither satisfied nor dissatisfied	Dissatisfied	Very dissatisfied	Total
Very unlikely	Count	0	3	7	23	14	47
	%	0.00%	1.60%	2.20%	1.90%	1.50%	1.70%
Unlikely	Count	8	27	23	133	92	283
	%	12.70%	14.80%	7.10%	11.00%	9.90%	10.50%
Neither likely nor unlikely	Count	7	26	57	134	91	315
	%	11.10%	14.30%	17.60%	11.10%	9.70%	11.60%
Likely	Count	23	58	123	469	356	1029
	%	36.50%	31.90%	38.10%	38.90%	38.10%	38.00%
Very likely	Count	25	68	113	447	381	1034
	%	39.70%	37.40%	35.00%	37.10%	40.80%	38.20%
Total	Count	63	182	323	1206	934	2708
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Detailed analysis by respondents' level of knowledge of the legal industry

Table 46 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') varied across their levels of knowledge of the legal industry, with respondents with a poorer legal knowledge being the least likely to think that the solicitor in the scenario was totally responsible for what happened. The results of a test we performed to compare the responses we collected across our respondents' levels of knowledge of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 46: Responses to Q1 (responsibility for what happened) by respondent's level of knowledge of the legal industry

		Strongly disagree	Disagree	Neither agree nor disagree	Agree	Strongly agree	Total
Not responsible	Count	3	17	9	18	0	47
	%	1.10%	1.20%	0.60%	2.00%	0.00%	1.10%
	Count	24	107	74	46	3	254
	%	9.20%	7.50%	5.10%	5.00%	2.60%	6.10%
Somewhat responsible	Count	40	306	254	158	16	774
	%	15.30%	21.40%	17.50%	17.20%	14.00%	18.60%
Mostly responsible	Count	80	475	437	255	28	1275
	%	30.70%	33.30%	30.10%	27.80%	24.60%	30.60%
Totally responsible	Count	114	522	678	439	67	1820
	%	43.70%	36.60%	46.70%	47.90%	58.80%	43.60%
Total	Count	261	1427	1452	916	114	4170
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 47 shows that the responses our respondents provided to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') varied across their levels of knowledge of the legal industry. Respondents with a better legal knowledge were more likely to think that outside factors were totally responsible for what happened than respondents with a poorer legal knowledge. Respondents with a poorer legal knowledge, however, were more likely to think that outside factors were mostly responsible than respondents with a better legal knowledge. The results of a test we performed to compare the responses we collected across our respondents' levels of knowledge of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 47: Responses to Q2 (extent to which outside factors were responsible) by respondent's level of knowledge of the legal industry

		Strongly disagree	Disagree	Neither agree nor disagree	Agree	Strongly agree	Total
Outside factors not responsible at all	Count	35	180	223	156	24	618
	%	13.40%	12.60%	15.40%	17.00%	21.10%	14.80%
Outside factors slightly responsible	Count	81	419	456	317	31	1304
	%	31.00%	29.40%	31.50%	34.60%	27.20%	31.30%
Outside factors somewhat responsible	Count	58	461	473	244	30	1266
	%	22.20%	32.40%	32.60%	26.70%	26.30%	30.40%
Outside factors mostly responsible	Count	67	309	236	142	17	771
	%	25.70%	21.70%	16.30%	15.50%	14.90%	18.50%
Outside factors totally responsible	Count	20	54	61	56	12	203
	%	7.70%	3.80%	4.20%	6.10%	10.50%	4.90%
Total	Count	261	1423	1449	915	114	4162
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

and Table 48 shows that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') varied only marginally across their levels of knowledge of the legal industry. Respondents with a poorer legal knowledge were slightly more likely to think that the solicitor in the scenario would be very unlikely to behave in the same way in the future than respondents with a better legal knowledge. Respondents with a better legal knowledge, however, were slightly more likely to think that the solicitor in the scenario would be very likely to behave in the same way in the future then other respondents. The results of a test we performed to compare the responses we collected across our respondents' levels of knowledge of the legal industry confirmed that the differences we observed were not statistically significant (see Annex B).

Table 48: Responses to Q3 (likely to behave in the same way in the future) by respondent's level of knowledge of the legal industry

		Strongly disagree	Disagree	Neither agree nor disagree	Agree	Strongly agree	Total
Very unlikely	Count	122	563	580	380	52	1697
	%	47.80%	40.20%	40.50%	42.10%	45.60%	41.40%
Unlikely	Count	81	511	456	294	28	1370
	%	31.80%	36.50%	31.90%	32.60%	24.60%	33.40%
Neither likely nor unlikely	Count	13	106	155	56	7	337
	%	5.10%	7.60%	10.80%	6.20%	6.10%	8.20%
Likely	Count	26	150	167	114	16	473
	%	10.20%	10.70%	11.70%	12.60%	14.00%	11.50%
Very likely	Count	13	69	73	59	11	225
	%	5.10%	4.90%	5.10%	6.50%	9.60%	5.50%
Total	Count	255	1399	1431	903	114	4102
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 49 shows that the responses our respondents provided to Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') varied across their levels of knowledge of the legal industry. Respondents with a better legal knowledge agreed more with statement than respondents with a poorer legal knowledge. The results of a test we performed to compare the responses we collected across our respondents' levels of knowledge of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 49: Responses to Q4 (extent to which events could have been prevented) by respondent's level of knowledge of the legal industry

		Strongly disagree	Disagree	Neither agree nor disagree	Agree	Strongly agree	Total
Strongly disagree	Count	1	9	11	8	0	29
	%	0.40%	0.60%	0.80%	0.90%	0.00%	0.70%
Disagree	Count	5	17	18	12	3	55
	%	1.90%	1.20%	1.20%	1.30%	2.60%	1.30%
Neither agree nor disagree	Count	10	40	46	24	2	122
	%	3.80%	2.80%	3.20%	2.60%	1.80%	2.90%
Agree	Count	61	434	357	201	13	1066
	%	23.40%	30.50%	24.60%	22.00%	11.40%	25.60%
Strongly agree	Count	184	924	1019	670	96	2893
	%	70.50%	64.90%	70.20%	73.20%	84.20%	69.50%
Total	Count	261	1424	1451	915	114	4165
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Table 50 shows that the responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') varied across their levels of knowledge of the legal industry. Respondents with a better legal knowledge were more likely to take the complaint further and report the solicitor to the regulator of solicitors than respondents with a poorer legal knowledge. The results of a test we performed to compare the responses we collected across our respondents' levels of knowledge of the legal industry confirmed that the differences we observed were statistically significant (see Annex B).

Table 50: Responses to Q5 (likelihood of reporting to the SRA) by respondent's level of knowledge of the legal industry

		Strongly disagree	Disagree	Neither agree nor disagree	Agree	Strongly agree	Total
Very unlikely	Count	7	21	25	16	5	74
	%	2.70%	1.50%	1.70%	1.80%	4.40%	1.80%
Unlikely	Count	41	198	116	82	10	447
	%	16.00%	14.00%	8.10%	9.00%	8.80%	10.90%
Neither likely nor unlikely	Count	33	204	168	87	7	499
	%	12.90%	14.50%	11.70%	9.60%	6.10%	12.10%
Likely	Count	92	565	562	349	37	1605
	%	35.90%	40.10%	39.30%	38.40%	32.50%	39.00%
Very likely	Count	83	422	560	374	55	1494
	%	32.40%	29.90%	39.10%	41.20%	48.20%	36.30%
Total	Count	256	1410	1431	908	114	4119
	%	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%

Our results showed that overall respondents did not attribute responsibility for the potential misconduct of a solicitor in the hypothetical scenario we used in our survey differently depending on the perceived ethnicity or gender of the solicitor who provided the service. Respondents were not more or less likely to attribute responsibility to outside factors (such as bad luck or unforeseeable circumstances) as opposed to factors within the control of the solicitors when the service was provided by Asian or Black solicitors than they were when the service was provided by White solicitors. Neither were they more or less likely to attribute responsibility to outside factors when the service was provided by female compared to male solicitors.

Respondents were also not more or less likely to believe that the solicitor who provided the service could have prevented what happened when the service was provided by White solicitors than they were when the service was provided by Asian or Black solicitors, or when the service was provided by female or male solicitors. And, although some respondents indicated that Asian and Black solicitors, as well as male solicitors, may be more likely than White and female solicitors respectively to behave in the same way in the future (in relation to the potential misconduct), they reported no difference in the likelihood about whether they would report the solicitors to the SRA for potential misconduct.

Our results, however, provided evidence that respondents with different characteristics attributed responsibility in different ways.

• The ethnicity of respondents affected how they attributed responsibility for what happened. White respondents ascribed responsibility for what happened to the solicitor in the scenario more than Asian or Black respondents (see table 11). There were also some differences by place of birth (see table 16), although these characteristics did not affect the likelihood the respondents would report solicitors to the SRA for potential misconduct (see table 15 in relation to ethnicity and 20 in relation to place of birth).
• Female respondents were more likely to attribute responsibility for what happened to the solicitor (see table 21), but male respondents were slightly more likely to believe outside factors were mostly responsible (see table 22), although there was no significant difference between the genders in terms of the likelihood to report solicitors to the SRA for potential misconduct (see table 25).
• Older respondents were more likely to attribute responsibility to the solicitor in the scenario than younger respondents (see table 26) and less likely to attribute responsibility to outside factors (see table 27). But older respondents were not more or less likely to report them to the SRA for potential misconduct (see table 30).
• Respondents not in paid work (e.g., homemaker, retired or disabled) and unemployed respondents were respectively the most and the least likely to attribute responsibility to the solicitor in the scenario (see table 31). Employment status also affected the likelihood to report solicitors to the SRA for potential misconduct (see table 35).
• Respondents with different levels of experience of, satisfaction with, and knowledge of the legal industry attributed responsibility differently.

Taken together, the results of our survey provided evidence that respondents attributed responsibility for potential misconduct differently. Different attributions, however, seemed to be linked more to respondents' socio-demographic characteristics and levels of experience of, satisfaction with, and knowledge of the legal industry than to the ethnicity or gender of the solicitor who delivered the service.

Scenario used in the survey

You were recommended a solicitor for advice about taking your neighbour to court after they caused damage to your property. The solicitor said they were really busy, and were often working late, but would be able to take on your case and would fit the work in around other cases. You provided them with a lot of background information about the dispute, including medical evidence from your GP about how it has affected you. The solicitor took some files home one evening, including yours, to work on them. The files were left in their car overnight. During the night, their car was stolen. The car and your files were recovered a couple of days after, but the solicitor warns you that the thief may have read your file and the personal information it contains. A friend has told you that you can complain about solicitors to the regulator of solicitors and law firms, such as the Solicitors Regulation Authority.

Names and pronouns used to indicate ethnicity in the survey

• Mrs Sunita Kumar (Asian-female),
• Mr Sukjunder Singh (Asian-male),
• Mrs Mariam Namagembe (Black-female),
• Mr Anthony Olukayode (Black-male),
• Mrs Mary Jones (White-female),
• Mr Andrew Clarke (White-male).

We have set out below the results of the tests we applied to identify where the differences highlighted in our analysis were statistically significant.

Table 1 shows that the responses we collected were statistically different across gender-ethnicity combinations only for Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?'). For all other questions, we could not rule out the possibility that the differences we observed were attributable to chance.

Table 1: Comparison between responses across solicitor's gender-ethnicity combinations

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	4.883	3.149	48.243	6.38	8.118
df	5	5	5	5	5
Asymp. Sig.	0.43	0.677	<.001	0.271	0.15

Table 2 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') and Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') differed across respondents of different ethnicities in a statistically significant way. The responses to Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?') and Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') differed across respondents of different ethnicities in a marginally statistically significant way. The responses to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?') did not differ across respondents of different ethnicities in a statistically significant way.

Table 2: Comparison between responses across respondents' ethnicities

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	21.563	9.917	10.273	78.120	4.177
df	4	4	4	4	4
Asymp. Sig.	<.001	.042	.036	<.001	.383

Table 3 shows that the responses our respondents provided to Q1 to Q4 differed across respondents born in different places in a statistically significant way. The responses our respondents provided to Q5 (i.e., 'You complained to [solicitor name] … but you were not happy with their response. How likely would you be to take the complaint further and report [solicitor name] to the regulator of solicitors?'), instead, did not vary across respondents born in different places in a statistically significant way.

Table 3: Comparison between responses across respondents' places of birth

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	15.493	26.412	31.483	46.913	5.031
df	6	6	6	6	6
Asymp. Sig.	.017	<.001	<.001	<.001	.540

Table 4 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?') differed between female and male respondents in a statistically significant way and that the responses our respondents provided to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') differed between female and male respondents in a marginally statistically significant way. The responses to the other questions did not differ statistically between female and male respondents.

Table 4: Comparison between responses across respondents' genders

	Q1	Q2	Q3	Q4	Q5
Mann-Whitney U	2050209.500	2137668.500	2039221.500	2134753.500	2091202.500
Wilcoxon W	4103560.500	4445694.500	4285361.500	4182029.500	5
Z	-3.520	-.877	-1.930	-1.245	-.974
Asymp. Sig. (2-tailed)	<.001	.380	.054	.213	.330

Table 5 shows that the responses our respondents provided to Q1(i.e., 'To what extent do you think [solicitor name] is responsible for what happened?'), Q2 (i.e., 'To what extent do you think factors outside of the control of [solicitor name] (such as bad luck or unforeseeable circumstances) are responsible for what happened in the scenario?'), and Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') differed across ages in a statistically significant way. The responses to the other questions did not differ statistically across ages.

Table 5: Comparison between responses across respondents' ages

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	295.063	166.049	1.396	196.811	1.527
df	2	2	2	2	2
Asymp. Sig.	<.001	<.001	.498	<.001	.466

Table 6 shows that the responses our respondents provided to Q1, Q2, Q4, and Q5 differed across their employment statuses in a statistically significant way. The responses to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') on the contrary did not differ across our respondents' employment statuses in a statistically significant way.

Table 6: Comparison between responses across respondents' employment statuses

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	76.440	37.022	2.266	62.551	16.327
df	5	5	5	5	5
Asymp. Sig.	<.001	<.001	.811	<.001	.006

Table 7 shows that the responses our respondents provided to Q1, Q2, Q4, and Q5 differed across our respondents' levels of experience of the legal industry in a statistically significant way. The responses to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') instead did not differ across levels of experience of the legal industry.

Table 7: Comparison between responses across respondents' levels of experience of the legal industry

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	175.709	71.830	2.140	81.334	10.251
df	2	2	2	2	2
Asymp. Sig.	<.001	<.001	.343	<.001	.006

Table 8 shows that the responses our respondents provided to Q1 (i.e., 'To what extent do you think [solicitor name] is responsible for what happened?'), Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') and Q4 (i.e., 'To what extent do you agree that [solicitor name] could have prevented what happened?') differed across levels of satisfaction with the legal industry in a statistically significant way. The responses to the other questions dis not vary across levels of satisfaction with the legal industry in a statistically significant way.

Table 8: Comparison between responses across respondents' levels of satisfaction with the legal industry

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	20.033	2.638	13.912	29.200	7.372
df	4	4	4	4	4
Asymp. Sig.	<.001	.620	.008	<.001	.118

Table 9 shows that the responses our respondents provided to Q1, Q2, Q4, and Q5 differed across levels of knowledge of the legal industry in a statistically significant way. The responses to Q3 (i.e., 'How likely do you think it is that [solicitor name] will behave in the same way and leave documents in his/her car in the future?') on the contrary did not differ across levels of knowledge of the legal industry in a statistically significant way.

Table 9: Comparison between responses across respondents' levels of knowledge of the legal industry

	Q1	Q2	Q3	Q4	Q5
Kruskal-Wallis H	52.758	22.781	5.702	28.338	63.171
df	4	4	4	4	4
Asymp. Sig.	<.001	<.001	.223	<.001	<.001