THE POSSIBILITY OF WELLBEING : PRELIMINARY RESULTS FROM SURVEYS OF AUSTRALIAN PROFESSIONAL LEGAL EDUCATION STUDENTS

Surveys of students in the Australian National University’s Legal Workshop Professional Legal Education (PLE) program showed that students began and ended a core course with low levels of psychological distress. In contrast to other studies, we found no evidence of elevated symptoms of depressive symptoms and no signs that this PLE program may impair wellbeing. While small increases in average stress symptoms were observed, this was associated with a more positive course experience. Selfperceptions of professional identity and concurrent legal employment predicted lower distress, suggesting that wellbeing is enhanced by building meaningful connections between legal education and students’ wider identities.


INTRODUCTION
Although the quantitative study of psychological distress in Australian law students and lawyers has only been a recent endeavour, the findings to date have been consistent in their disheartening results.High levels of distress, particularly in the form of symptoms associated with depression, have been found in a troublingly large proportion of law students. 1The invariance of these findings is also concerning.Very similar results were obtained across different institutions, years of study, teaching and learning environments and student demographics.Similar findings about high levels of mental health concerns in lawyers would seem to imply that this trend continues unabated after law school and into the profession. 2 Despite this accumulating evidence about law students and lawyers, little is known about the psychological wellbeing of Professional Legal Education (PLE) students -the component of legal education that takes place in Australia after the completion of a four to five year undergraduate (LLB) or three-year graduate (JD) law degree in order to meet the prerequisite requirements in order to be admitted as a legal practitioner in any Australian jurisdiction. 3Indeed, one might be forgiven for assuming that the levels of psychological distress among PLE students would be just as high as LLB/JD students and lawyers, given the trajectory which begins in law school (pre-PLE) and continues into practice (post-PLE).
Such an assumption should not be left untested.Understanding the patterns of psychological distress and wellbeing in PLE students is important not only to fill in the gap in the research evidence, but also because of the unique characteristics in terms of pedagogy, student characteristics and motivations.The PLE context can offer much to inform the broader research project on psychological wellbeing in legal education and practice, as well as the efforts to develop educational or practice environments which allow students and lawyers to thrive both in their study or practice as well as in their lives generally.

II THE PLE CONTEXT: CONTINUITIES AND DISCONTINUITIES
PLE provides a very different context to the LLB and JD environment.As its name suggests, Professional Legal Education is designed to prepare law students to make the transition to practice.Over a relatively short period of time, PLE encourages the metamorphosis of identity from law student to legal professional.This focus on the practice of law necessarily requires a pedagogical pivot away from the emphasis on appellate case law and theory in the law degree.Nonetheless, most PLE in Australia takes place within a higher education framework.Students still take part in formal content-based instruction, are set assignments, assessed, graded and receive feedback along the way.There is a continuation of some familiar features of law school, while other aspects to which students have become socialised over the past three to five years are intentionally removed.In some Australian jurisdictions, particularly in the past, PLE (also known as PLT: practical legal training) could be undertaken in the form of articled clerkships, but this has already been phased out or is not particularly prevalent as a structured higher education program has been the dominant pathway to admission to practice in Australia.
QUT Law Review Volume 14, Number 1, 2014 Special Edition: Wellness for Law 29 PLE therefore involves what we term a 'pedagogy of disruption', nurturing of a new identity while requiring at least some dissolution of familiar ways of thinking, relating and acting created by law school. 4The simultaneously familiar and unfamiliar terrain of this new stage of legal education needs to be studied.Do PLE students adapt to this change in teaching and learning?Or does this new and different way of learning and preparing for practice cause or exacerbate psychological distress?
Understanding the psychological wellbeing, experiences and expectations of PLE students is also important because of its position at this critical transition point on the journey from law student to lawyer.If PLE students already experience high levels of psychological distress even before entering practice -before joining a profession whose norms and practices can be hostile to wellbeing -then this does not augur well for future members of the profession, or the profession as a whole.It is incumbent on PLE providers to ensure that they are not the cause of further distress.Answering this question is an empirical task, requiring quantitative data to monitor levels of psychological distress or specific areas of concern, and to respond in the curriculum appropriately.
PLE also provides an all-too-brief window of opportunity to develop awareness of psychological wellbeing and to build resilience and strengths for not only the immediate course of study, but for the career about to begin.This kind of effective preparation for practice goes beyond mere knowledge of the law.It requires understanding the systemic causes of distress in practice and the obstacles to wellbeing which affect lawyers' ability to be healthy, ethical and professional practitioners.Accordingly, as PLE develops students' skills and competencies for practice, it must simultaneously address issues of psychological wellbeing.Here, too, thorough ongoing evaluations of the program are needed so that any reforms and interventions are based on the best possible evidence.

A ANU Legal Workshop
It is, however, very difficult to speak about PLE in a generic sense.The Australian PLE landscape is less homogeneous in pedagogy than Australian LLB and JD programs.Each  to complete these transactions to a competent standard in small teams mentored and supported by experienced current legal practitioners (Practice Mentors).An intentionally small student-to-mentor ratio of around 4:1 is enforced to provide students with ample support while they develop and demonstrate their skills and develop professional relationships and identities. 6e integrated simulated practice environment of the PPC is designed to promote active, self-determined learning, collaboration, error and discovery.Students reflect on and respond to ongoing formative feedback to achieve competency rather than individually perusing pre-set materials and readings to construct a definitive answer for (often) arbitrary marks. 7The PPC replaces the standard classroom model with a team-based learning environment where substantive law, personality (and, sometimes strong personalities), and legal procedure (both formal and informal) play out over real time.In this environment, negotiations, compromises, ethical challenges, conflict and wellbeing arise in both spontaneous and planned ways and are as important as the substantive law.
Initially based on the curriculum created by the Glasgow Graduate School of Law for their PLE program,8 the practice management curriculum is the 'glue' within the PPC.It binds the learning of substantive areas of law with the skills and attitudes needed to transform this knowledge into successful, collaborative professional practice.In the latter half of 2012 the practice management component of the curriculum was substantially revised to address psychological wellbeing explicitly and throughout the program.Based on the core conviction that wellbeing cannot be separated from professionalism, values and ethics,9 the revisions sought to embrace the theoretical and empirical research on the positive role that ignorance and uncertainty can and should play in legal education and practice, as well as the power of providing students with the skills to take action on their values by using the 'Giving Voice to Values' (GVV) curriculum.
The practice of law is an environment saturated with uncertainty, yet lawyers find themselves as part of a legal system which is imbued with the language of certainty, rules, precedents and clear-cut results.Identifying, naming and managing uncertainty has become a cross-cutting theme within the PPC.Part of the change to the practice management curriculum has been about encouraging our students to learn how to be comfortable with recognising and acting positively when confronted with the uncertainties of the legal system.10 To complement this focus on uncertainty and professionalism, the GVV curriculum has been adapted to the online PLE.The aim is to take students' ethical conflict-management skills beyond mere thought processes and theoretical challenges to a set of skills that will enable them to act on their values. 11Based initially in business ethics, this curriculum is founded upon research on the role of rehearsal in forming habits that can influence change in situations of ethical conflict.The GVV curriculum complements the broader goals of practice management by providing a practical framework for mentors to assist students to deal with conflict constructively, create high-functioning, high-trust learning teams and assist in difficult practice contexts such as negotiations with colleagues and clients.
These pedagogical reforms have been implemented cautiously but thoroughly integrated throughout the whole curriculum of the PPC in order to avoid the strategy of only introducing remedial interventions or relying on compartmentalised modules to address psychological wellbeing as a discrete topic.Indeed, the depth of the reforms continues to alarm some traditionalists who hold a preference for a didactic method of training students to 'think like a lawyer' based on argumentative and rational skills alone.
With such a major change to curriculum, we wanted evidence that students were benefiting and not being harmed by what we are doing.We also wanted an opportunity to test two criticisms of ANU's PLE model.Firstly, does a predominantly online learning environment increase isolation and therefore psychological distress?Secondly, does the simulated practice environment, with its occasional unpredictability but constant workload, result in impaired wellbeing, especially in a population of students who are likely to be working full-time in equally busy and demanding jobs?These and other questions required an empirical study.While some of these questions are yet to be answered, we present here some preliminary results from a multi-year, multicohort study that began after the curriculum reforms had been introduced.

A Design and Psychometric Instruments
The survey methodology and design that we used was based on earlier research on first-year ANU LLB and JD students conducted by Molly Townes O'Brien, Kath Hall and Stephen Tang in 2009 and 2010 (ANU First Year Study).We created an online survey which was completed by PPC students at two time points: at the beginning (within the first week) of the PPC and at the end of the 18-week course (a busy and usually stressful time for students).Like the ANU First Year Study, the survey contained a number of well-validated self-report psychometric scales addressing psychological wellbeing and distress, as well as other individual differences measures.
The 21-item version of the Depression Anxiety Stress Scale (DASS-21) was used as the measure of psychological distress.12Items on each of the three DASS-21 subscales (depression, anxiety and stress) are derived from non-overlapping clinical symptoms. 13Participants rate the extent to which they have experienced the symptom over the past week on a four-point scale, ranging from 'Did not apply to me at all' to 'Applied to me very much, or most of the time'.
It is important to note that the names of the DASS-21 subscales refer to specific clinical features which are slightly different from the ordinary meanings implied by their names.As such, the depression subscale primarily addresses symptoms such as loss of self-esteem and incentive, feelings of worthlessness and inertia.These symptoms are commonly experienced in a major depressive episode, which is more than transient low mood.The anxiety subscale refers to the heightened physiological reactivity, situational anxiety, fear and panic (in a whole-body sense) which are common across the cluster of otherwise quite disparate anxiety disorders.Finally, the stress scale measures the presence of tension, irritability, difficulty concentrating and overreaction to external stressors. 14The DASS-21 meaning of 'stress' therefore refers to a more specific and acute set of symptoms than is understood through the everyday meaning of the word. 15ores on the DASS-21, which range from zero to 21 on each of the subscales, can be transformed into a set of five severity categories (normal, mild, moderate, severe and extremely severe). 16These categories refer to the intensity and frequency of symptoms rather than the severity of a diagnosable disorder.17Just because a person scores within the 'severe' range on the depression subscale does not necessarily mean that she or he has 'severe depression'.Rather, this result is better interpreted as meaning that the frequency or intensity of depressive symptoms is severe, suggesting the likely presence of a psychological disorder. 18s it is wholly inappropriate to make clinical diagnoses based on the DASS-21 alone, 19 these categories therefore should only be used as a rough guide.Nonetheless, a general rule of thumb is that scores in the moderate category and above indicate a level of psychological distress which is significantly affecting the person's life, relationships and daily activities.
In addition to having good psychometric properties for a brief instrument, the use of the DASS-21 is now established in the Australian law student wellbeing context.The use of the same scale allows the results of this survey to be compared directly with both the 2009-10 ANU First Year Study and the findings obtained by Larcombe and colleagues at Melbourne Law School in their 2011 study of LLB and JD students (the MLS Study). 20ntinuing with the methodology of the ANU First Year Study, the 5-item Satisfaction with Life Scale (SWLS) was included as a positive measure of psychological wellbeing. 21Similarly, in the beginning-of-course survey, students were asked a set of 18 questions about their expectations of and motivations towards the PPC.At the end of the course, these questions were rephrased in terms of actual experiences.While these data have been discussed elsewhere, 22 a subset of eight questions (those relating to positive course-related or personal development expectations and experiences) was used to form a simple measure of positive expectations or experience about the PPC. 23 in the ANU First Year Study, participants could elect to receive a confidential follow-up telephone call from an independent psychologist if their DASS-21 scores were at a level which indicated the possible presence of a psychological disorder or significant impairment with everyday life.The purpose of the phone call was not to offer formal counselling but to explore options for help and support which were available to the student.This was important given that many ANU Legal Workshop students are physically located in other states, territories and countries and may not have access to on-campus counselling services.The psychologist was equipped with information with which to assist students in finding appropriate services in the student's local area.Information about free telephone counselling services (eg, Lifeline) and online information and self-help programs (eg, e-hub) 24 were provided to all participants at the beginning and end of the surveys.
One addition to the methodology was the use of a de-identified participant code in both surveys.Participants generated an alphanumeric code based on snippets of personal information which was easily recalled but could not be feasibly used to identify the participant.This enhanced the repeated-measures design of the survey methodology by allowing participants' responses from the two surveys to be matched while maintaining participants' anonymity.Importantly, this meant that individual changes over the course could be measured where participants participated in both surveys and accurately entered their participant code.
Participants could also opt-in to be surveyed again after their admission to practice, and again approximately 12 months later as part of a related Transition to Practice project. 25

B Procedure and Participants
A successful pilot test was run in the first half of 2012 with 53 and 49 participants at the beginning and end of the course, respectively.The survey was then deployed in the PPC course in the second half of 2012.A total of 388 students at the beginning of the PPC, and 245 students at the end of the course, agreed to participate in the research component.This represents a response rate of approximately 75 percent for the initial survey and 50 percent for the end of course survey.This is an acceptable response rate which provides a good degree of statistical power for the analyses described below. 26l PPC students were invited to take part in the surveys, which were embedded into a broader qualitative and quantitative reflective self-assessment task at the beginning and end of the course.In addition to the optional psychometric questions, this reflective task also contained a number of short-answer questions and questionnaires about their working styles, approaches to conflict, dealing with ethical problems and their perceptions about professionalism and the qualities of a 'good' team.The questions and reflections on the survey questions were the basis of a web conference with their virtual team and practice mentor at the beginning and end of the course.
Students were told about the dual purpose of the task before they commenced the survey.Students were asked to work through the survey and at least think about the questions as part of their reflective task, even if they did not wish to participate in the research by submitting their responses.At the end of the survey, students were then given the option to opt-in to the research project.This was completely voluntary, and students' decision to, or not to, participate in the research component was unknown to their mentors or other teaching staff.
Students' responses on the reflective questions were saved in an online database but were only accessible by the student themselves, unless the student had elected to participate in the research component.Practice mentors and other teaching staff had no access to students' individual responses.This research methodology was reviewed and approved by the ANU Human Research Ethics Committee.

A Demographics
At the beginning of the PPC, the mean (average) age of the participants was 28 years of age, although the median was lower, at 25 years of age.The mean being higher than the median is explained by the 'long tail' of a smaller number of older students.Approximately 60 percent of students were female, with 39 percent male, and just under 1 percent of participants not identifying as either female or male.Just under two-thirds (64%) of the participants were enrolled in the GDLP on a full-time basis.
There was considerable diversity in the number of institutions at which participants completed their LLB/JD, with over 30 universities represented.Most participants transitioned from their law degree to PLE within a short period of time.Forty-two percent of participants completed their LLB/JD within the past six months and 75 percent within the past 12 months.Almost nine out of ten (89%) of participants were employed, with 62 percent in full-time work when they started the PPC.Of the students who were working, 70 percent were employed in a law-related capacity (eg, as a paralegal, law clerk or legal policy officer).Almost half (45%) of the total sample of 338 students were working fulltime in a law related position.
Most participants intended to become a legal practitioner upon completing their PLE (which qualifies them for admission to practice).Overall career intentions were measured using a single-item bipolar visual analogue scale scored from zero to 100, where zero represents a definite intention to work in a non-law-related career and 100 represents a definite intention to work in a law-related career.At the end of PPC survey,27 29 percent of participants definitely intended to work in a law-related field (100/100).Half of all participants expressed a very strong intention to practice law (at least 85/100).Only 15 percent of participants were more inclined to consider a non-law career (scoring less than 50 on the scale).

B Depression, Anxiety and Stress: DASS-21 Results
The results from the DASS-21 can be analysed in several complementary ways.The first perspective is to look at the results on the Depression, Anxiety and Stress subscales at both the beginning and end of the PPC both as categorical (severity rating) and dimensional (mean score) data.Next, we can compare the results from this sample of PPC students with data from the ANU First Year and MLS Studies.Finally, we turn to individual changes in levels of psychological distress and their predictors at an aggregate level.Table 1 shows the percentage of students in each of the DASS-21 severity categories for each subscale.The first and most noticeable observation is that more than three-quarters of the PPC students were in the 'normal' category on all three subscales in both surveys.Together with the comparative results shown in Figure 1, this finding is a good reminder (before going deeper into the results), that the majority of law students do not show concerning levels of psychological distress throughout their studies. 28though the DASS-21 severity scales are useful, and will be revisited when making comparisons with other Australian studies, we wanted to compare our data against more precise normative measures (a representative sample of people with whom our sample of PLE students can be validly compared). 29On measures of psychological distress, are our PLE students as a population distinguishable from a random sample of people in the Australian community?If so, how do they differ?
To test this, the mean (average) scores on each DASS-21 subscale were compared with Australian community population norms, which were split into two age categories: 18-24 and 25-90 year olds. 30Splitting the normative data into two age groups makes sense because of the elevated levels of psychological distress during young adulthood which decline and plateau into adult life. 31However, the width of the latter age group (covering more than six decades) is a potential problem.As most of our PLE students are just on the cut-off between the two age 28 See also Figure 1. 29 This can be considered as the statistical analogue to the 'ordinary person' test, though less fictitious. 30 categories; it may be more accurate to compare them to their younger peers than to community members more than double their age.Better comparative Australian norms for the DASS-21, however, are unavailable.As such, the comparisons may err on the side of over-estimating the level of distress in our sample of PLE students.

Depression Subscale
Turning to the depression subscale of the DASS-21, the overall pattern was that PLE students at the beginning and at the end of the PPC were not distinguishable from their community peers.That is, PLE students did not show an elevated level of depressive symptomology at either the beginning or at the end of the PPC.In fact, younger (18-24 year old) PLE students at the beginning of the PPC had levels of symptoms which were significantly lower than their age-matched peers (Table 2).There was also no statistically significant change in the frequency or severity of depressive symptoms over the PPC.Moreover, end of PPC students were no more likely to be in a higher severity category on the DASS-21 subscale compared with beginning of PPC students (Table 1). 33These findings stand in sharp contrast with previous studies, where the depression subscale accounted for the highest proportion of distress and showed the steepest rate of increase.In comparison, while students in the first few weeks of law school showed in the ANU First Year Study low levels of depressive symptoms (79.5% in the normal range, 11.7% moderate or above), their scores on the depression subscale increased significantly over the year (Figure 1(a)).By the end of the year, only 54.9 percent were in the normal range and 31.5 percent of students were in the moderate or above categories.Similarly, 27.3 percent of students in the MLS Study had DASS-21 depression scores in the moderate, sever or extremely severe categories.This substantial increase in distress, in terms of vulnerability towards depression, was pleasingly absent in our sample of PLE students.

Anxiety Subscale
On the anxiety subscale, there were once again no statistically significant changes over the PPC.Similar to the depression subscale, end of PPC students were no more likely to have a higher level of categorical severity on this subscale. 34ounger (18-24 year old) PLE students had levels of anxiety symptoms which were indistinguishable from the community normative sample (Table 2).For students aged 25 and above, however, their mean anxiety subscale scores were slightly elevated compared with the normative sample.This was observed both at the beginning and end of the PPC.However, this finding is subject to the caveat above about the arbitrary age cut-off in the normative sample.
Previous studies indicated that LLB and JD students started off with a higher than normal level of anxiety symptoms which remained constant over the course of their studies (Figure 1(b)).In contrast, ANU PLE students had much lower baseline scores on the anxiety subscale (15.7% and 17.5%), even though their scores were still slightly higher than what would be expected in the wider population.This relatively low level of distress was sustained over the duration of the course.
Compared with the ANU First Year and MLS Studies, there was also a much smaller proportion of students in the severe and extremely severe categories of the anxiety subscale (Figure 1).Nonetheless, anxiety symptoms account for the most distress in terms of the three subscales, with just over 15 percent of students both at the beginning and end of the PPC showing concerning levels of anxiety symptoms (Table 1).We emphasise that, once again, there was no evidence that the PPC had any causal effect on students' anxiety symptoms.However, the comparatively high background levels of anxiety symptoms, which seems to be invariant across different studies, may warrant more attention.

Stress Subscale
The one component of the DASS-21 in which there were clear differences between the beginning and end of PPC surveys was the stress subscale.Students' end of PPC stress subscale scores were significantly higher than their scores at the beginning of the PPC.Their end of course scores, but not beginning of course scores, were also higher than the normative sample for both age categories (Table 2).End of PPC students were approximately twice as likely as beginning of PPC students to have stress scores in a higher severity category. 35 such, the possibility that the PPC may have contributed to this rise in stress symptoms cannot be discounted, as will be explored further below.On the other hand, these results must be considered in context.There were fewer PPC students who had stress subscale scores in the moderate and above categories (8-13%) compared with the ANU First Year (16-20%) and MLS students (25%).In addition, the largest shift in the severity categories was from the normal to the mild categories (the latter increasing from 6.2% to 11.8%; Table 1), with a slightly smaller increase in the severe category (from 1.8% to 6.5%).The overall size of the change was small (a mean difference of 1.13 points, on a 21 point scale), especially given the low baseline scores.By contrast, a similarly-sized increase in stress subscale scores was also observed in the ANU First Year Study (Figure 1(c)), although this was of greater concern given that this change occurred on top of higher baseline stress scores at the very beginning of law school.

C Satisfaction with Life
Results from the SWLS provided convergent evidence to the DASS-21.As expected, subjective life satisfaction was negatively correlated with the DASS-21 at both time intervals. 36That is, as life satisfaction increases, psychological distress decreases, and vice versa.Similar to the overall stability of the DASS-21 scores from the beginning to the end of course, there were also no statistically significant changes in SWLS scores across the PPC.PLE students' perceptions of their life satisfaction remained at the uppermost bounds of the average range. 37his result reinforces the view that PLE students begin and end the course with a good level of life satisfaction, with nothing in the results to suggest that the course adversely affects this.

D Individual Changes and Course Experience
So far, these analyses have looked at whether students at the beginning and end of the PPC, as two distinct groups, differ on their average levels of psychological distress.We now turn to a more fine-grained approach by looking at whether individual students became more (or less) psychologically distressed over the course.
These analyses are based on a smaller subset of the participants.Only 65 percent of the final sample (157 out of 245 participants) could be matched to their responses at the beginning of the PPC.Despite using a matching algorithm which could tolerate and correct a small number of errors, the matched set was smaller than expected. 38One likely reason was due to participants entering different codes or not following instructions.Additional verification measures have been included in more recent iterations of the survey.Nonetheless, the current dataset provides a sufficient sample size to explore individual-level change, at least for the purposes of reporting some preliminary findings.
Each student's end of PPC DASS-21 scores was compared with her or his scores at the beginning of the PPC.The results were very similar to the group-level analyses described above.As shown in Table 3, there were no statistically significant changes on scores on the individual depression and anxiety subscale scores over the PPC.There was, however, a small but statistically significant increase in the stress subscale, which accounts for the small overall increase in general psychological distress.These findings indicate that students who had high levels of psychological distress at the end of the course were more likely to have started the course with high levels of distress.It also means that the majority of students who started the course with low levels of distress were also likely to finish the course with low levels of distress.On average, only a one-point increase in their DASS-21 stress score was observed.
To explore further whether individual changes in psychological distress might have been influenced by the course itself, we looked at the relationship between simple changes scores on the DASS-21 (ie, subtracting the end of PPC scores from the beginning of PPC scores) and the measure of positive course experience.If students were distressed by the PPC, then we would expect that positive DASS-21 change scores (ie, change scores greater than zero, indicating that students became more distressed over the course) would be negatively correlated with positive course experiences.That is, we would expect students to be less likely to report that the course was engaging, interesting, and helpful for professional development if over the same period of time they became significantly more psychologically distressed.However, no such finding was observed.There was second code.For example, the Levenshtein distance between the participant codes 4HI8JUNO2 and 4HI8JUCA2 is 2, as the code differs by two letters.Code pairs with Levenshtein distances ≤ 4 were identified as possible matches, based on the inference that the same participant entered the code differently either for either survey.Possible matches where the participant's age (or age + 1) and gender were not identical for both surveys were discarded.Finally, the remaining code matches were compared manually according to their constituent alphanumeric parts, prioritising parts of the code which were more likely to be entered incorrectly. 39Paired samples t-tests were used to establish whether the difference between the means was statistically significant.The difference in the means between the two survey groups is statistically significant if its t-value exceeds a certain threshold given the degrees of freedom (df) calculated from the pooled sample size.In these analyses, df = 156.Statistically significant differences, using a 95% confidence level, are indicated by asterisks (*) in the last column.Statistical significance does not necessarily imply practical significance.
no association between overall distress (GPD) and positive course experiences. 40or did changes in anxiety or depression subscale scores predict positive course experience. 41This means that whether a student had a positive experience of the PPC had nothing to do with their levels of anxiety or depressive symptoms.
Interestingly, however, changes in the stress subscale score were positively correlated with a positive course experience. 42In other words, students whose stress scores increased over the semester (which was not uncommon) also tended to report a more positive experience of the PPC.This finding suggests that mild levels of increased stress may not necessarily be interpreted as harmful or debilitating. 43One student that provided feedback in an unsolicited email after the course, described this dialectic in the clearest possible terms: '… the past 18 weeks have been the most stressful of my life.And I've loved every minute of it.What an amazing experience.' 44 The presence of manageable stress can also signal a constructive sense of disruption through learning.As Townes O'Brien and Tang have suggested, the study and practice of law necessarily involves struggle and contact with negative experiences.However, if such struggle is appraised as meaningful and is backed up with intrinsic motivations and a supportive learning environment, it need not be something to be avoided. 45

E Predictors of Psychological Distress
We can therefore set aside any assumptions that PLE students necessarily continue along the trajectory of psychological distress as found in LLB and JD students and lawyers in practice.Our sample of PLE students showed a much more healthy profile of psychological wellbeing compared with their Australian LLB and JD peers.Nonetheless, it remains the case that around 10-15 percent of our students do experience levels of psychological distress which is concerning.The next question, therefore, is to explore what personal and situational factors predict possible vulnerabilities to depression, anxiety disorders or other mental and psychosocial problems.Or, to invert the question: are there resilience factors which insulate PLE students against problematic levels of distress?
At the beginning of the PPC, before any substantive work was undertaken, age and gender were statistically significant but small-scale predictors of elevated levels of depressive symptoms. 46Younger students were more likely to have 40 r = .086,p = .286.Correlations for subscale scores were also near zero and not significant. 41Depression: r = .072,p = .369;Anxiety: r = .020,p = .800. 42Zero order correlation: r = .210,p = .008.Hierarchical linear regression: b = .265,p = .006;R 2 adj = .029,controlling for age, gender, depression and anxiety change scores, and concurrent law work (see below). 436 Age: b = -.121,p = .029;Gender: b = -.146,p = .009.R 2 adj = .022.Due an extremely small cell size, participants who identified as neither female nor male (n = 2) could not be included in these analyses.However, this is likely to be a population at particular risk of high levels of higher depression scores, as expected, although there was no effect on anxiety and stress scores.Contrary to expectations informed by psychiatric epidemiology literature, 47 male students had higher levels of depressive symptoms than female students. 48On the other hand, female students had a slightly higher stress subscale scores at the beginning of the PPC. 49No gender differences were observed on the anxiety subscale. 50 the end of the PPC, age and gender were no longer consistent predictors of psychological distress. 51Instead, other more robust predictors emerged from factors relevant to the connection between the student and her or his situational context, namely law-related work experience and legal professional identity.Looking at a subset of the 89 percent of end-of-PPC students who were employed, those who were working in a law-related capacity had significantly lower scores on all three DASS-21 subscales compared with students who were working in a job unrelated to law. 52For instance, the mean depression subscale score for students who were working in a law-related job was 2.38 while the mean score for students working in a non-law-related job was 4.17. 53This difference was statistically significant, controlling for age and gender.Similar results were found for the anxiety 54 and stress subscales. 55In other words, there was something unique about work experience in a law-related capacity which insulated against distress.To flip this around, students working in a job which was not connected with their studies were more prone to experience higher levels of psychological distress.The strength of this finding was unexpected and deserves further investigation.However, the present survey instrument did not ask about the nature or amount of work undertaken.We suspect that these may be important mediating variables.A univariate analysis of variance (ANOVA) was used to test whether the type of work had a significant effect on DASS-21 scores: F(1,216) = 9.992, p = .002,η 2 p = .045. 54Law work mean: 1.77, non-law work mean: 3.13; F(1,216) = 8.355, p = .004,η 2 p = .038. 55Law work mean: 5.01, non-law work mean: 6.42; F(1,216) = 5.107, p = .025,η 2 p = .024.
The extent to which students had a coherent sense of identity as a legal professional also emerged as a robust insulator against distress at the end of the PPC. 56This sense of professional identity increased over the PPC (from 52:100 to 58:100 on a single-item measure, with higher scores indicating a greater sense of identity as a legal professional). 57Even after controlling for legal work experience, 58 higher levels of identity as a legal professional also predicted lower levels of distress on the depression, 59 anxiety, 60 and stress subscales. 61Taken together with the protective role of concurrent legal work experience, this suggests that a sense of congruence between work experience, the teaching and learning environment of PLE, and emerging professional identities are all important to promote and sustain psychological wellbeing.

F Summary
The results from the DASS-21 indicate that a sizeable majority of ANU PLE students began their core PPC course with levels of psychological distress at or around levels expected of Australians in the general community.Apart from a slight elevation in anxiety symptoms in students aged over 25 (which may be attributable to how the age categories of the normative sample were defined), there were low levels of problematic psychological distress both at the beginning and end of the PPC.As a group, our PLE students did not stand out from the 'ordinary person' in terms of psychological distress.
These findings are in sharp contrast with the patterns of elevated psychological distress, particularly depressive symptoms, as observed in the ANU First Year and MLS Studies.Our results also show that if the LLB and JD law school environment does systematically contribute to psychological distress, its effects do not seem to have a longer term deleterious effect.Three-quarters of the survey participants completed their LLB or JD within the past 12 months, but DASS-21 scores at the commencement of the PPC were not substantially different to 'normal' levels.
This pattern of overall low psychological distress was also stable across the PPC, despite it being an intensive period of learning through simulated practice.At both group and individual levels, there were no changes in levels of depression or anxiety symptoms on the DASS-21.A small but statistically significant increase on the stress subscale was observed, although this was found to be positively correlated with positive experiences of the PPC in terms of learning, professional development, and preparation for practice.Students who were working in a law-56 Rachael Field, James Duffy and Anna Huggins, 'Supporting transition to law school and student well-being: the role of professional legal identity' (2013) 4 International Journal of the First Year of Higher Education 15. 57 This increase was statistically significant: t(581) = 2.655, p = .008,CI. 95 =[.014, .093].At the same time, law student identity declined from a mean of 58/100 to 48/100 (t(581) = -4.142,p < .001,CI .95= [-.149,-.053]). 58As expected, students who were working in a law-related capacity had a greater sense of legal professional identity (61/100 compared with 52/100, t(243) = 3.081, p = .002,CI. 95 =[.032, .149]),but professional identity still had an independent effect. 59b = -.196,p = .019,R 2 adj = .052.This model (and those below) also controlled for age and gender. 60b = -.207,p = .014,R 2 adj = .032. 61b = -.204,p = .016,R 2 adj = .028.related job, and those who felt like they had a well-formed identity as a legal professional, were far less likely to report high levels of distress.The congruence and interplay between life experiences and the teaching and learning environment may therefore act as a resilience factor.

G Methodological Limitations
In all voluntary surveys, there is a risk that the data could be biased if there are systemic reasons why some people decided to participate and others did not.This risk is partially mitigated by the satisfactory response rate in this study, aided by embedding the survey as a voluntary component of a broader reflective task within the coursework.There was a substantial drop in the response rate for the end of PPC survey, although the stability of the aggregate results compared with the individual change data suggests that the sample remained sufficiently representative.Empirical research, however, does not rest on a single set of results.As such, replication efforts are underway with more recent and larger PPC cohorts.
Looking at individual changes over the PPC also allowed us to take a step closer towards seeing whether the course has a causal effect on psychological distress, compared with just looking at group-level differences.However, as will be discussed below, there are many possible influences on distress levels well beyond the PPC itself.The individual change results were also based on a smaller subsample of students, although the results were consistent with group-level findings.
Methodologically, one key limitation is that we were again focused on the negative measurement of psychological distress rather than the positive measurement of psychological wellbeing.These two things are not the same.However, given the absence of data about psychological distress in PLE students, we thought it was important to begin with a risk-averse approach and first assess for negative effects.We anticipate that future surveys will shift towards assessing the role of positive attributes, values and psychological needs without completely disregarding the assessment of psychopathological symptoms.
The interval between the two surveys (18 weeks at most) was also shorter compared with the one academic year in the ANU First Year Study. 62On one hand, this may have been an insufficient amount of time to measure the onset of sustained psychological distress, rather than acute 'stress' (in the everyday sense).This should be balanced against the teaching and learning environment of the PPC, which is designed to be a concentrated period of formative, immersive learning.
A cautious approach is needed when attempting to generalise from our findings to the Australian PLE experience.This is particularly so given the dissimilarity of teaching and learning approaches adopted by the different institutions across the country.Further research is needed, especially with other PLE students in different institutions.Only with such data can we start to disentangle the different influences on psychological wellbeing at this stage of legal education.

V IMPLICATIONS FOR LEGAL EDUCATION AND WELLBEING RESEARCH
Despite these caveats, we can make a bold but affirming claim from the interim results of our surveys: not all legal education is inherently distressing and harmful to students.The results from our surveys show that it is possible to learn about substantive legal content, the processes of lawyering, the nature of being a lawyer, the interpersonal dynamics and the role of personal values and strengths in the practice of law without also exhibiting signs of impaired mental health, particularly in the form of depression or problematic anxiety.
We can be confident from these results that the PPC is not directly causing psychological impairment for students.Placed in contrast with the less encouraging findings for LLB and JD students, we have reason to suspect that the teaching and learning environment of the PPC may help to maintain low levels of baseline psychological distress through the promotion of supported autonomy and competence, which in turn rewards students for being intrinsically motivated.
PLE is unlike the law school curriculum of the LLB or JD.In our program, the PPC's pedagogical framework intentionally moves away from encouraging students to become competent through 'thinking like a lawyer' in an excessively rational, 63 self-degrading 64 and experience-erasing 65 way. 66In fact, the PPC actively tries to undo some of these ways of thinking and problem-solving which are not advantageous for the real-world practice of law.Specifically, the 'hidden curriculum' of adversarialism in the law school environment 67 is replaced by the need to work cooperatively with fellow team members, virtual supervisors and virtual opponents.Web conferences and ongoing written communications with experienced practitioner mentors provide ongoing feedback (and feed-forward) as well as a constant source of connectedness, leaving no room for isolated learning. 68The uncertainties of practice are acknowledged and named such that they become an everyday part of studying and practicing, rather than something to be feared in the quest for a 'right answer'. 69 the same time, there is a degree of fragility and unpredictability in the PPC's emphasis on naturalistic and experiential learning.professional development, but often in ways which are not entirely foreseeable or controllable.However, such experiences can also be transformed, with guidance from mentors and team members, into useful learning experiences about working with others under pressure.This appears to be successful so far in the PPC, and may partially explain the relationship between mild increased stress and positive course experience.For example, a student, when reflecting on the role of uncertainty in the law stated that: [G]roup/peer consultation [in the PPC] was effective in calming the nerves and making me realise that I'm allowed to be uncertain about things, and moreover, that not everyone knows everything or knows all of the rules -I'm not behind just because I don't know the answer to something, as long as I have methods for finding it out; I'm not wrong because I don't format something in the best way, I just need to use my common sense to make a call and if it's wrong, learn from that and move on, and so on.It also reminds me that it's a normal part of being a lawyer to question yourself and to question things around you -a constant state of uncertainty makes you who you are, and without it, you'd be less cautious, less measured, and probably less of a good lawyer as a result. 70e findings about the importance of concurrent law-related work experience in lowering psychological distress are noteworthy, especially given the reality that PLE study is often undertaken on top of existing work commitments.The PPC is designed around the reality that most students are employed and often working in a legal capacity.It is also assumed that a large majority of students will proceed almost immediately into professional practice, an assumption supported by the results.Within this specific context, our results suggest that the PPC environment may empower students to apply what they are learning in the course in their 'live' work environment as well as in their simulated practice.The seamless transfer of experiences, skills and acquired processes from study to work seems to be beneficial in terms of developing competence and enhancing the meaning of both study and work.That this opportunity can only be seized to its full potential by those students who are working in a law-related job may explain why only legal work experience insulates students against depression, anxiety and stress.
The content of the PPC curriculum may also be directly benefiting students in their immediate work situations.This may be particularly the case in relation to the GVV exercises with students coached about how to be assertive about valuedrive decisions; where to take such a stance may appear to be difficult or, euphemistically, a 'career limiting move'.For instance, one exercise involves the student preparing what they would say to their supervising partner when they wish to refuse to work over a weekend, knowing that the assigned work is not time-critical and they have not had a weekend off in over a month.Being confident and competent to have these difficult conversations has direct consequences for psychological wellbeing.
An alternative explanation is that legal work experience may prepare students for the disjunctions between PLE and law school.Working in a legal context may prepare students for the unique and unexpected features of the practice of law of which they will also encounter in their simulated working environment.Such exercises may include finding precedents, dealing with the courts, and working with clients and other lawyers.Simulated practice would be made easier with the benefit of real-life experience.One student reflected on the uncertainties experienced in the PLE in these terms: I think the law degree teaches a lot of skills but it is a bit lacking in the 'soft skills' required once in the workplace.This can make new graduates feel very uncertain.
[In the PPC,] I am working towards learning better and more confident communication and skills for the workplace that will reduce anxiety and uncertainty when a task is given to me. 71th of these explanations suggest that law schools may have a good reason to encourage students early in their studies to seek out law-related employment. 72ince this is not always possible, promoting clinical legal education would be another way for students to develop this experience. 73However, it is not merely the work experience that is beneficial, but the integration between legal work and the content of the curriculum.Legal education could be made to be less distressing to the extent that courses recognise these 'outside' experiences as being integral to students' learning.In the PPC, students are encouraged to bring all their resources and experiences (wherever and however obtained) to the table when completing tasks and reflections.Free to apply experienced gained outside the formal study of law, students are often delighted to discover that they knew how to do things which they did not realise they knew (so-called 'unknown knowns'). 74e confluence between learning how to practice law and real-life experiences from related concurrent work provides opportunities for mutual autonomous support.This connection between different domains is likely to assist in the development of competence, confidence and overall psychological wellbeing, as predicted by self-determination theory (SDT). 75Recognising and making use of these connections also encourages the development of meaningful professional identities, even if it involves a degree of struggle and stress along the way.Future surveys will include SDT measures to test this hypothesised relationship in more detail.
Finally, for the wellbeing research project in general, there is a negative and a positive conclusion.Our results, similar with previous studies on this point, suggest that law schools and the profession may need to do more to address the problematic levels and forms of anxiety in students and lawyers.There is no evidence suggesting that (professional) legal education causes higher anxiety symptoms.However, we could do more to recognise and support the 15 percent or so of students who have significantly elevated anxiety symptoms throughout their studies.To do so would be consistent with a recent public health focus on anxiety and anxiety disorders, which has so far have taken a back seat to depression despite their high (and possibly rising) prevalence. 77e literature has recognised that law school can encourage patterns of thinking and relating which may increase vulnerabilities to depression.However, the effect of law school in intensifying or activating existing vulnerabilities associated with anxiety has not been explored in much detail.Problematic anxiety in the law school context could be expressed in the form of vigilant symptoms, such as excessive worry and perfectionism, constant physiological arousal (eg, 'butterflies in the stomach', feelings of nausea or panic) and a constant 'scanning' of the environment for perceived threats, with harmful consequences for relationships, self-esteem and competence. 78Anxiety is also commonly reflected in unhelpful forms of avoidance, trying to stay away from the source of stress, which may be at odds with learning objectives especially in more performative tasks (eg, moots, presentations, simulated practice and clinical legal work).
In terms of what this means for course design, it may not be that the solution is to remove all stresses from a course, but instead to recognise and acknowledge the stress and/or anxiety points in a course and provide the appropriate structures and supports to encourage sustainable ways through these stress points.For example, as a student wrote about the PPC: I think that this course is very indicative of the stress and workload of a workplace.The stress and pressure feels real and it is very important to remove the uncertainty of feeling you have a huge workload by making a plan.I learnt about planning and prioritising tasks and got some good advice on how to manage workloads, how to establish the most important tasks and how to work smarter.I also learnt the importance of asking questions to reduce uncertainty and stress. 79rning to the positive and wellness-oriented aspect of the research and pedagogy, our survey results also act as an exhortation to look at the broader context of legal education and practice.If, as we have said, we wish to encourage students to bring their 'whole self' to law school and practice, then we must also recognise that this whole, unified self (as SDT recognises) involves much more than their law school self.PLE students spend a large proportion of their time at full-time work, relationships and family commitments.These things can, of themselves, contribute to psychological wellbeing or distress.While these results may reflect the more flexible structure of PLE programs, we suspect that the lives of LLB and JD students would not be too dissimilar.This, however, is an opportunity for further empirical research to complement the thorough but proximal data on demographic, participant-related and course-related influences on law student wellbeing. 80e influence of non-law-related study (especially for combined degree LLB students), work, relationships, social activities, as well as the rich tapestry of joys, discoveries, excitements, disappointments and sorrows of early adulthood are all relevant to wellbeing and the prevention of distress.There is a tendency for law schools to elevate the 'law student' identity above others.We do not doubt that the law student identity is important and rightly cherished by most students. 81owever, emphasising its primacy may also unconsciously reinforce some of the harmful and stereotypical aspects of law school, such as its functional utility for gaining entry into an elite, powerful and profitable profession.This could potentially be mitigated by validating the importance of other identities and experiences which coalesce in ways which are consistent with intrinsic motivations and personal values beyond the instrumental purposes of law school. 82Admittedly, this may be easier to do in a PLE context where law student identity is already waning, overtaken by legal professional identity. 83cognising the value of other identities is perhaps best done not through another set of activities or programs, but by stepping back.The same may apply for other wellbeing-oriented projects.There is an opportunity for legal education at all levels(even for the ANU's GDLP in the midst of ongoing curriculum reform), to embrace a healthy balance between active reform and a via negativa: intervention by intentional omission. 84The attitude of 'doing more', either in terms of 79 See participant journal entry above n 700.This journal extract was from a different student in the summer 2013 PPC course. 80See especially Larcombe and Fethers, above n 1. 81 Larcombe, Tubaga, Malkin, et al, above n 1, 420 (70% of JD and 50% of LLB students agreeing with the statement 'I really like being a law student'); Townes O'Brien and Tang, above n 45, 3 (65% of first year LLB/JD students agreeing with the statement 'Studying law contributes to my personal growth'; 77% agreeing with the statement 'Studying law is a good decision'). 82Townes O'Brien and Tang, ibid 6-9. 83 It is therefore not so easy to attribute psychological distress directly to legal education alone.Our results suggest that a meaningful congruence between different identities, roles and experiences can create conditions under which psychological needs can be met, facilitating wellbeing, motivation and achievement.We observed this in the case of legal work which co-occurs with simulated practice and reflective learning.The positive correlation between legal professional identity and low levels of psychological distress is another important connection which needs further study.
Our results show that that PLE, and legal education more generally, has the capacity to achieve its pedagogical and professional outcomes without impairing wellbeing in the process, not that a given curriculum or set of interventions will guarantee a clean bill of psychological health.There is not necessarily a continuation of a trajectory of unwellness from law school into legal practice.It is possible for PLE students, in this important preparatory and transitory stage, to maintain low levels of depression, anxiety symptoms and even thrive under the mild stress of a challenging learning environment.
We therefore end up with a more complex, but rightfully complex, picture of the relationship between psychological wellbeing and legal education.However, it is this same complexity which allows us to break away from the expectation that we study law students in order to find distress.The task ahead is more exciting and imbued with hope: first to replicate these findings, then to explore other wellbeing-relevant connections and finally to apply the lessons from the research back to the curriculum and beyond.This endeavour is made only easier by the realisation that to achieve this, we need only begin by listening to and learning from our students who are well and thriving: the overwhelming majority of them.

2
Norm Kelk et al, Courting the blues: attitudes towards depression in Australian law students and legal practitioners (Brain & Mind Research Institute, 2009) 42; Christopher Kendall, 'Report on psychological distress and depression in the legal profession' (Law Society of Western Australia, 2011); Beaton Research and Consulting and Beyondblue, 'Mental health in the workplace' (2011); Frances Gibson, 'Psychiatric disability and the practising lawyer in Australia' (2012) 20 Journal of Law and Medicine 391. 3

Table 3 : Individual change scores on DASS-21 subscales. 39 DASS-21 Subscale Mean change
E-mail received 28 May 2013.Thank you to this student for giving permission for this quote to be used anonymously. 45Molly Townes O'Brien and Stephen Tang, 'Stop the struggle for the struggle's sake: make it meaningful' (Paper presented at the 2013 International First Year in Higher Education Conference, Wellington, 8 July 2013) <http://fyhe.com.au/past_papers/papers13/5A.
pdf> ; see also Mark D Seery, Raphael J Leo, Shannon P Lupien, Cheryl L Kondrak and Jessica L Almonte, 'An upside to adversity?moderate cumulative lifetime adversity is associated with resilient responses in the face of controlled stressors' (2013) 24 Psychological Science 1181.
Team work is not uncommonly a source of stress and conflict.Similarly, a supported environment where mistakes are tolerated, if not encouraged, can facilitate deep learning and 76 71 This journal extract was from a different student in the summer 2013 PPC course. 72Cf Leanne Mezrani, 'Feeling the pressure', Lawyers Weekly (online), 13 February 2013 <http://www.lawyersweekly.com.au/features/feeling-the-pressure>. 73See, for example, Colin James, 'Seeing things as we are: emotional intelligence and clinical legal education' (2005) 8 International Journal of Clinical Legal Education 123; Tony Foley, Margie Rowe, Vivien Holmes and Stephen Tang, 'Teaching professionalism in legal clinic: what new practitioners say is important' (2012) 17 International Journal of Clinical Legal legal education on law students: a longitudinal test of self-determination theory' (2007) 33 Personality and Social Psychology Bulletin 883; Kennon M Sheldon and Lawrence S Krieger, 'Does legal education have undermining effects on law students?evaluatingchanges in motivation, values, and well-being' (2004) 22 Behavioral Sciences and the Law 261; Krieger, above n 9.76Townes O'Brien and Tang, above n 45.
Cisler and Ernst HW Koster, 'Mechanisms of attentional biases towards threat in the anxiety disorders: an integrative review' (2011) 30 Clinical Psychology Review 203; for a general overview of anxiety disorders, see Beyondblue, Understanding anxiety (2013) <https://www.bspg.com.au/dam/bsg/product?client=BEYONDBLUE&prodid=BL/0384&type =file>.QUT Law Review Volume 14, Number 1, 2014 Special Edition: Wellness for Law 50 But note the role of developing a positive professional identity (as one part of identity) from the very beginning of law school: Field, Duffy and Huggins, above n 56; Kath Hall, Molly Townes O'Brien and Stephen Tang, 'Developing a professional identity in law school: a view from Australia' (2009) 4 Phoenix Law Review 21. 84 As conceptualised by Nassim Nicholas Taleb, Antifragile: how to live in a world we don't understand (Allen Lane, 2012) 301.This should not be confused with the subtractive nature of QUT Law Review Volume 14, Number 1, 2014 Special Edition: Wellness for Law 51 curricular interventions or expectations on students, may sometimes come at the expense of the deliberate spaces and gaps needed for integrative identity formation and values-oriented discovery.Admitting this is, perhaps, the most disruptive of all pedagogies.