Researching judicial ethical codes, or: how to eat a mille-feuille?

Judicial ethical codes have become increasingly important in member states of the European Union as a point of reference for judges and as a means of securing public trust in judiciaries. In the ENCJ’s survey on the independence, accountability, and quality of judicial systems in Europe, sub-indicators were included to measure the existence of judicial ethical codes and relating training and supervision. This survey and the critical scholarly discussion about it provide valuable lessons for further empirical research on the perceptions of judges in EU member states relating to professional-ethical codes and standards. This article explores the most important lessons for this envisaged research. The main methodological considerations which come to the fore are: 1) to clearly delineate the research aims and central terms which are used in the survey; 2) to avoid insularity by including a representative and diverse group of respondents; 3) to validate and contextualise the results through a multi-method approach and critical debates with scholars and stakeholders.


Introduction
Ethical (or deontological) codes for legal professionals have been compared to a mille-feuille, the French pâtisserie composed of layers of puff pastry and pastry cream and topped with powdered sugar or icing. 1 Indeed, the metaphor seems to work when considering the formal 'crumbling' and substantive heterogeneity which the set of available codes in the world displays. Crumbling, firstly, can be seen in the variety of codes on good conduct of legal professionals, the actors involved in developing and establishing these codes, and the different categories of legal professionals to which these codes apply. Written ethical codes for inter alia judges and lawyers have proliferated in the 21st century, both in national systems and internationally. 2 These codes are developed within the professions, but increasingly also with the involvement of governments and legislators. 3 Heterogeneity, secondly, concerns the variety in the content of ethical codes for legal professionals. Differences exist with regard to the specificity of codes, their guiding or binding nature (sometimes specified in the used terminology, e.g. code, guide, guiding principles), and the clarity of provisions. In this landscape of ethical codes, the guiding force for legal professionals and the internal coherence of norms are hampered by overlaps between codes as well as the inconsistent distribution of norms over internal guidelines and more formal types of regulation. 4 Despite the proliferation of ethical codes for judges in Europe, not much research has been conducted on the content of these codes and their influence on judicial functioning. Yet, professional ethics constitute one of the core elements of judicial culture, meaning "the features that shape the way in which the work of a judge is performed and valued within particular legal systems". 5 In this regard, judicial ethical codes are an important factor for achieving goals set by the European Commission for realising effective judicial cooperation in the European Union (EU) in the framework of the Lisbon Treaty. 6 Against this background, this article aims to present a reflection on the methodological considerations for an envisaged empirical research on the perceptions *Elaine Mak is Professor of Jurisprudence at Utrecht University and a researcher at the Montaigne Centre for Rule of Law and Administration of Justice. This article connects with a research project on European judicial culture(s), funded with a Vidi grant from the Netherlands Organisation for Scientific Research (NWO). Thanks to Erin Jackson, Niels Graaf, and Frans van Dijk for commenting on a draft version of this article and to the anonymous reviewers for their useful feedback. Any mistakes remain my own E-mail: e.mak@uu.nl of judges concerning the influence of judicial ethical codes in the EU 7 In this reflection, lessons will be drawn from the design and results of the survey on judicial independence and accountability, which was conducted by the European Network of Councils for the Judiciary (ENCJ) in 2017. 8 This survey and critical analyses from scholars 9 will be consulted to answer the central research question for this article: which methodological considerations should be kept in mind when elaborating an empirical research design for studying the influence of judicial ethical codes in EU member states?
First, the topic of judicial ethical codes and its relevance in the EU context are further introduced (2). Following this, a brief overview of judicial ethical codes in selected EU member states will highlight main points of attention for a comparative and empirical research on these codes (3). Next, relevant parts of the ENCJ survey and a draft design for the envisaged new study on judicial cultures will be described as two examples of research design on judicial ethical codes, of which the combined results will be able to clarify the connections between formal arrangements for national judiciaries and the realisation of de facto independence and accountability (4). Finally, an analysis will be made of the relevant methodological considerations for researching judicial ethical codes, taking into account lessons which can be drawn from the ENCJ survey and the critical discussion of this survey (5). Some concluding remarks finalise the analysis (6).

Reasons for researching judicial ethical codes in the EU
Those who like the mille-feuille will know that it can be quite hard to eat this pastry gracefully. 10 Likewise, finding one's way in the plurality of ethical codes for the legal professions and dealing with substantive issues of clarity and coherence of norms can be a challenge for legal professionals themselves as well as for members of the wider society. Yet, tackling this challenge is of high importance for addressing changed societal demands. Transparency and accountability have become central values for ensuring public trust in professional performance, also for judiciaries, influenced by theories of new public management. 11 Besides this, independence remains a prominent value for professionals and it is currently understood first of all as the professional's neutrality with regard to (political) power. 12 These observations ring true also for the specific category of legal professionals formed by judges. Judiciaries are grappling with societal and organisational demands of effective, efficient, and client-oriented dispute settlement. 13 At the same time, the independent position of judges vis-à-vis political powers requires constant maintenance in systems which strive to realise the ideal of the rule of law, understood here as the prevention of the arbitrary use of power. 14 Against this background and with the aim of securing public trust, ethical codes for judges have been developed in many countries, in particular during the past two decades. 15 Indeed, ethical codes can be a 'sign board' for society, which confirms that the core values of the society are taken seriously by judiciaries and individual judges, and these codes can express a set of concrete norms which may serve as a basis for accountability. 16 A further role of ethical codes emerges in the context of the EU, where national judges have responsibilities regarding the uniform application of EU law and the guarantee of equal legal protection for citizens. Yet, developments such as the rule-of-law crises in Poland and Hungary 19 underline the societal and political challenges for realising cooperation based on mutual trust and recognition, for example with regard to the European arrest warrant in criminal cases. 20 In this context, ethical codes have the potential of providing a normative common ground and concrete guidance for judges in EU member states to decide on professional-ethical dilemmas relating to the values of effective and efficient judicial protection and the realisation of the rule of law. In this regard, comparative influences from other EU member states and from European and international codes could inspire the content and strengthen the persuasive force of national ethical codes in specific EU member states and in this way assist judges in claiming professional neutrality vis-à-vis political actors in their country. 21 For judicial ethical codes to fulfil their functions properly, in particular to ensure access to justice and public trust in judiciaries, a sound formal structure and internal coherence of norms are required. 22 Still, knowledge on the substantive similarities and differences between national judicial ethical codes in the EU and on the influence of these codes on judicial functioning is scarce. 23 Pertinent questions for research then are: how have core values for judicial functioning, such as independence and accountability, been presented in judicial ethical codes in EU member states? How do judges perceive the influence of these codes for guaranteeing the core values of judicial functioning?

Judicial ethical codes in the EU: a mille-feuille indeed
A brief overview serves to demonstrate the diversity in judicial ethical codes in the EU with regard to the included core values and intended function of the code. 24 In the overview in this section, a selection is presented of codes from four EU member states: the United Kingdom (England and Wales), France, Denmark, and Romania. These member states represent different legal traditions (common law and civil law), 25 different professional cultures (consensus and honour), 26 and different degrees of realisation of the ideal of the rule of law (established and 'new' liberal democracies). 27 Based on this overview (3.1), some main characteristics of judicial ethical codes in Europe, which can serve as a starting-point for the design of empirical research, come to the fore (3.2).

A tour d 'horizon
In England and Wales, a Guide to Judicial Conduct was first published in 2004 and updated several times since. 28 This guide was established by a working group of the Judges' Council and it has been authorised by the Lord Chief Justice of England and Wales and the Senior President of Tribunals. 29 The guide applies to judges, coroners, and magistrates and it mentions three basic principles: independence, impartiality, and integrity. 30 In the introduction to these principles, a distinction is made between the terms 'guide' and 'code' in order to emphasise the non-binding nature of the listed ethical norms. The purpose and status of the Guide are explained as follows: It is based on the principle that responsibility for deciding whether or not a particular activity or course of conduct is appropriate rests with each individual judge.
This Guide is therefore not a code, nor does it contain rules other than where stated. Instead, it contains a set of core principles which will help judges reach their own decisions.
In cases of difficulty or uncertainty, however, judges should always seek advice from the relevant leadership judge. 31 In France, the Conseil Supérieur de la Magistrature (CSM) established a deontological code for magistrates -encompassing judges as well as public prosecutors -in 2010. des magistrats. 33 Similar norms were developed for lay judges in the tribunaux de commerce for commercial cases 34 and the conseils de prud'hommes for labour cases. 35 In the elaboration of the code for magistrates, the CSM used comparative insights in order to connect with "a universal figure of the magistrate" that transpires from international legal instruments. 36 Furthermore, the CSM consulted citizens as well as judges during the process of drafting of the French code. 37 The French code lists six values for magistrates: independence, impartiality, integrity, legality, attention for the other, and discretion and prudence.
The French code specifies: "[t]he professional conduct of a magistrate cannot be left to his own discretion. It is determined by the law and it complies with ethical demands of its function." 38 The code is meant to provide magistrates with points of reference for their professional-ethical conduct. It also has the function of clarifying the complex nature of the functions of judges and public prosecutors to the legislative and executive branches of government, court personnel, and the general public. 39 Magistrates are reminded that ignorance of the core principles of independence and impartiality, granted to them by law, will compromise public trust. 40 Furthermore, magistrates should show by their integrity that they are worthy to decide on the exercise of fundamental rights by individuals. They are bound to the principles of honesty and loyalty and should by their actions confirm the supremacy of the law as well as the basic idea that justice is done in name of the French people. 41 In Denmark, the Association of Danish Judges has formulated Ethical principles for judges. These principles have been published in Danish and in English on the Association's website. 42 Taking into account inter alia the 2010 recommendation of the Council of Europe, the Association has formulated ten guiding principles distributed over four categories: independence, impartiality, and integrity, quality, openness, and accountability. 43 The Danish code was established with the aim of ensuring public trust. Furthermore, it responds to the Council of Europe's recommendation R (2010) 12, which provides that ethical principles of professional conduct "should be laid down in codes of judicial ethics which should inspire public confidence in judges and the judiciary". 44 The Danish code specifies that the listed principles "are a codification of that already applicable to judges and they are a supplement to the legislation's general rules on courts' and judges' affairs". 45 The Superior Council of Magistracy in Romania adopted a deontological code for judges and prosecutors in 2005. This code has been published in English too. 46 The code lists six values: independence of justice, promoting the supremacy of the law, impartiality of judges and prosecutors, exercise of professional duties, dignity and honour of the profession of judge or prosecutor, and activities incompatible with the judge or prosecutor position.
The Romanian code can be used in evaluations of the efficient functioning and integrity of judges and prosecutors by the legally competent bodies. 47 It contains detailed provisions, which give instructions to judges on how to act. As an example, the code regulates the interaction of the judiciary with politics in order to safeguard the value of judicial independence: Art. 4 -(1) In exercising their professional duties judges and prosecutors shall not be influenced by political doctrines.
(2) Judges and prosecutors must not militate in favour of other persons' adhering to a political party, must not participate in fund collecting for political parties and cannot allow the use of their prestige or image to such aims.
(3) Judges and prosecutors must not give any support to a candidate to a political type public function.

Main characteristics of judicial ethical codes
On the basis of the presented overview of judicial ethical codes in selected EU member states, some main characteristics of these codes can be identified. This overview also clarifies that the characteristics of these codes can be considered as sharing family resemblances rather than one common conceptual template. 48 The identified characteristics should be kept in mind in the design of empirical research on judicial ethical codes, as they clarify which stakeholders exist with regard to the codes and they highlight specific issues of content and effect of the codes. Moreover, this overview already highlights the importance of contextualisation of the outcomes of empirical research. The meaning and influence of judicial ethical codes in individual EU member states can only fully be understood when taking into account aspects of legal tradition, professional culture, and degree of realisation of the ruleof-law principle.
When considering judicial ethical codes as a specific category of norms, the following characteristics of these codes appear relevant for an empirical research: audience, content, type of norms, functions, and legal status.

Audience: judges and others
Ethical codes for judges apply to the active members of the profession. Some codes explicitly include retired judges. 49 Sometimes, related professions are addressed as well. A striking aspect of the French code is that it does not apply to lay judges, for whom separate codes were developed. In this regard, the French code's applicability to judges and public prosecutors relates to a specific elaboration of the rule-of-law principle in the constitutional status of magistrates and connected formal arrangements. 50 The applicability of the English guide of judicial conduct to coroners can be explained by the coroner's status as an independent judicial office holder, which has developed in the common law tradition. 51

Content: universal values?
The values expressed in the judicial ethical codes in the overview encompass the broad categories of independence, impartiality, professionalism, and integrity. The codes concern ethical conduct in the performance of professional tasks, such as conduct during court hearings or the deliberations on judgments. The codes also encompass broader conduct outside of the courtroom, for example political activity. Most of these codes have drawn inspiration from European and international documents, in particular the Bangalore Principles of Judicial Conduct developed under the auspices of the United Nations. 52 Still, national contextual aspects transpire, for example in the French code's emphasis on the dignity of the office of magistrate, which can be considered to reflect the professional culture of honour. 53

Type of norms: values and standards
Professional values present a frame of reference for proper judicial conduct. These values provide the normative underpinning for professional-ethical choices of judges. 54 In a different category, professional standards for judicial performance are minimum norms for the adequate handling of court cases. These standards are connected with the idea of 'judgecraft', which encompasses the minutiae of procedural arrangements in courts. 55 The content of these standards can range from court etiquette to break times or the processing of requests for adjournment of hearings and is intricately linked to experiences of procedural justice. 56 In comparison, the codification of standards has developed less than that of values. The distinction between values and standards is important for understanding the functions and effects of judicial ethical codes, as normative points of reference (values) have a different influence than more practical instructions (standards). Still, a caveat is in order. Although values and standards can be distinguished conceptually, overlaps could occur in the practical elaboration of norms for judicial conduct in codes and other guiding documents. For example, both values and standards can mention continued education as a means of maintaining expertise and professionalism for the judgment of cases. A risk of this kind of overlap is that professional standards are included in judicial ethical codes as objective 'measurable' norms, which enable sanctions against judges who fail to meet these standards rather than help protect their independence.

Functions
Judicial ethical codes can have different functions. Most codes mention the intended functions and distinguish between the guiding role for judges (internal function) as well as the explanatory role for society (external function). A further nuance is possible when we consider the dynamics in the internal organisation of judiciaries and the position of judges in this organisational context. dynamics should be taken into account when researching the influence of ethical codes from the point of view of individual judges. The possible functions of codes can then be conceptualised as follows: 57 -Motivating function: incentive for judges to perform better in the judicial office based on concrete norms for good conduct; -Accounting function: input for transparent and relevant disciplinary procedures, 'sign board' for society; -Educational function: input for the education and training of new judges (mostly inductively, e.g. through dilemma training); -Innovating function: space for adaptation of the judicial role to meet changing societal demands, based on a reflected solution of dilemmas.
As observed above, the accounting function becomes problematic if 'measurable' standards are included in judicial ethical codes. Then, the guiding nature of ethical norms as a means of strengthening judicial independence might be undermined by the use of these norms to sanction or fire judges.

Legal status: guidelines or binding norms
Judicial ethical codes can be presented as non-binding guidelines or as norms to which judges can be held accountable in internal evaluations or disciplinary proceedings. 58 In the overview, the status of non-binding general guidelines prevails in states where the rule-of-law principle is firmly embedded in the professional culture. In states where the rule-of-law principle is not yet as firmly embedded, for example in post-communist states in Central and Eastern Europe, judicial ethical codes more often appear to have a stronger binding nature and contain more specific norms. 59 This is visible in the example of the Romanian code, which encompasses specific and precise norms and explicitly mentions the role of the code in evaluations of judges and public prosecutors. 60 In empirical research, this distinction should be kept in mind, as it can be expected to colour the perceptions of judges with regard to the influence of ethical codes on their professional conduct inside the courtroom as well as in their interaction with political actors and society.

Research designs: two examples
In this section, two approaches to studying judicial ethical codes will be described, followed by an analysis of the relation of these research designs to the questions and aims of the respective projects. The first example concerns the ENCJ survey on judicial independence and accountability, in which indicators on judicial ethical codes were included to measure the objective accountability of judges and judiciaries in 23 member and observer countries to the ENCJ (4.1). 61 The second example concerns an empirical research project under development on the influence of judicial ethical codes in EU member states, which has the aim of assessing the possible alignment of professional-ethical values and practices of judges as a part of a European judicial culture (4.2). 62

ENCJ survey
The ENCJ's survey has several aims. 63 In the short run, the aim is to measure the de iure independence and accountability and perceived independence in countries in order to provide suggestions for the improvement of formal arrangements for the judiciary and judges. In the long run, the ENCJ aims to use the broad insights from the survey on the realisation of judicial independence in Europe to influence debates in countries where this principle is threatened, e.g. in Poland. Finally, the survey has the goal of giving a voice to judges, who can express their perceptions on the realisation of de facto judicial independence in their country.
The ENCJ's approach consists in the development of indicators and sub-indicators for judicial independence and accountability on the basis of: 1) an analysis of available international principles and experiences of judges in the ENCJ with regard to independence and accountability; and 2) the listing of actual arrangements to aspire to in relation to these principles. 64 The measurement through the survey is used to generalise outcomes in order to provide common aspiration levels and potential remedies for countries which do not yet meet these levels. 65 The ENCJ survey assesses formal independence and formal accountability of judges and judiciaries as well as perceived independence. The assessment of perceived accountability has not been included in the survey, presumably for practical reasons. 66 In the survey, judicial ethical codes are addressed as an indicator of the objective accountability of the individual judge. 67 This focus on accountability suggests that a connection is made to the function of ethical codes as a source of concrete norms and as a 'sign board' for society. 68 Objective indicators in this survey concerned "the legal and other objectively observable aspects of the legal system that are essential for independence and accountability". 69 The sub-indicators measured for judicial ethical codes were: the existence of a code of judicial ethics; if existent, the public availability of this code; the availability of training on judicial ethics; and the existence or not of a responsible body to provide judges with guidance or advice on judicial ethics. 70 For each of these sub-indicators, the developed questionnaire contained an option to answer with "yes" or "no". 71 The four sub-indicators on judicial ethical codes could together give a maximum score of 100%, with the result on each sub-indicator contributing 0% or 25% to the total score. The survey results show that scores on the four sub-indicators for judicial ethical codes yield a high average score of 85%. There is, however, a high variation between countries, with the lowest score being 0% and the highest score being 100%. 72 For this measurement of objective indicators, a standardised questionnaire was filled out by the judicial councils or, in countries where no judicial council exists, by other governance bodies. The information obtained through this self-evaluation could have been retrieved also in other ways, that is: it "can be checked by anybody who is knowledgeable about the legal systems concerned". 73 According to the ENCJ, the objective indicators "set a standard about how formal arrangements should look like. … For all indicators a high score is good and a low score bad." 74 As a general caveat, the authors of the ENCJ survey report warn that "the indicators have not been developed to create rankings of judicial systems, but can be used to discuss the strengths and weaknesses of judicial systems. Readers of the report are advised to treat the comparison of data from different countries with various geographical, economic and legal backgrounds with great caution." 75

Judicial cultures project
The judicial cultures project concerns an independent academic study on judicial cooperation in the EU, involving ethical, legal, and institutional aspects of judicial functioning. 76 The final aim of this research is to establish whether a shared normative framework for the functioning of national judiciaries in the EU -in other words: a 'European judicial culture' -can develop, which will enhance effective judicial protection and ultimately will contribute to the progressive realisation of the ideal of the rule of law in the EU. In the framework of this study, a survey is used to collect information on the perceptions which judges have about the functions and effects of judicial ethical codes in EU member states. These perceptions can include knowledge of and experience with codes as well as individual views on the role and usefulness of these codes. 77 A distinction is made between professional values for judicial conduct and professional standards for judicial performance. 78 This will enable assessing whether relevant differences between these categories can be identified, e.g. concerning the role of values and standards in external communication or in internal evaluation of judges, and the influence of codification (or not) of judicial ethical norms.
In this research, two hypotheses will be tested, which seem to reach further than the ENCJ's inquiry into judicial ethical codes as an indicator of objective accountability. 79 The first hypothesis is that judicial ethical codes and published professional standards can acquire a binding status through their use, e.g. through application in disciplinary procedures against judges or in internal individual evaluations of judges. The second hypothesis is that judicial ethical codes and published professional standards can be instruments which influence perceptions of de facto independence and accountability among judges or de facto independence and accountability as such (through a norm-setting effect).
The approach will be to develop indicators for professional-ethical performance of judges on the basis of: 1) a comparative analysis of the existence and content of judicial ethical codes and professional standards in selected EU member states; and 2) the listing of core elements of professional-ethical performance to aspire to on the basis of this comparative overview. The survey, which will receive a follow-up in interviews, aims to clarify how judges perceive these core elements and how they implement them in their daily practice. This analysis will enable drawing conclusions about the common ground (or not) between EU member states regarding the content of professional-ethical norms for judges and the ways of implementing these norms. Furthermore, this analysis will make it possible to assess the de facto effects of professional-ethical values and standards in realising judicial independence and accountability.
The research design for this study (see Box 1) presents a bottom-up approach to the development of indicators, which are drawn from the developed values and practices in national systems. This is different from the ENCJ's approach, which draws primarily from international principles. 80 Moreover, this research does not have the aim, as the ENCJ's project has, to improve judicial systems, criticise member states, or give voice to judges 81 Its central aim is to investigate whether alignment of professional-ethical values and standards in the EU, as a basis for effective judicial cooperation, can develop. 82 Of course, policy-makers, judicial networks and other stakeholders could use the results of this academic research to push their own agendas forward. Therefore, it will be important in the presentation of the results to emphasise the descriptive nature of the research and to distinguish its aim from the policy goals which stakeholders wish to address. As a follow-up to the survey and interviews with judges, project seminars will be used to present the results to a broader audience and to involve different stakeholders in the reflection on the results. Furthermore, the results of the empirical research in this project will be connected with a content analysis of judicial ethical codes in order to validate the results of the survey and interviews. Contextualisation will be realised through a more close-up analysis of EU member states, in which the elements of legal tradition, professional culture, and the degree of realisation of the rule-of-law principle will be integrated in the research.
Still, some methodological challenges for the empirical research design remain, e.g. the concern of ensuring representative and trustworthy responses to surveys and interviews and the concern of combining the measurement of indicators with richer contextual studies in an accessible format. In the next section, these concerns will be addressed.

Methodological considerations
Based on the analysis in the previous sections, some methodological considerations for empirical research on the influence of judicial ethical codes in EU member states can now be mapped. These considerations concern two phases. The preparations for research, firstly, encompass the following aspects: determining the object and aim of the research, developing a conceptual framework, using a multi-method approach (5.1). The conducting of the research, secondly, requires attention for the validity and the contextualisation of results of an empirical survey (5.2).
80 Most international principles were developed with input from national judges. Therefore, overlaps between national and international ethical codes can be expected to be found. 81 See supra, par 4.1. 82 See supra, par. 2.

Preparations: optimising the research design
The initial question, of course, is: what will be measured in a survey and to what end? In this regard, research on judicial systems is characterised by the tension between a policy-oriented and an academic outlook. 83 From a policy-oriented point of view, this research has the capacity to assist in addressing problems in the formal arrangements for judicial systems and problems in judicial functioning. From an academic point of view, more fundamental questions deserve attention as well, for example questions aimed at explaining specific aspects of judicial systems or questions aimed at supporting a normative reflection on the desired role of the judiciary in society. This more fundamental research does not always contribute to immediate problem-solving, but it can generate a better understanding of the foundations of judicial systems and insight into possible innovations for the future.
For the design of an empirical research, this tension between policy-oriented and academic outlooks entails that a choice is required.
Research aimed at problem-solving can make use of a 'light' theoretical framework, in which existing normative values for judicial functioning are described and ordered. At the same time, the limitations of this kind of research for rethinking the foundations of judicial systems should be acknowledged. 84 By contrast, fundamental research aimed at understanding or at stimulating normative debate can dive deeper into the interpretation and contextualisation of norms for judicial functioning, while setting more modest goals in terms of its contribution to legislative and policy reforms.
Against this background, the two presented research designs have complementary objects and aims. The ENCJ survey has the benefit of yielding a large set of objective data, which could signal 'red flags' -meaning a situation which requires urgent attention 85 -in the realisation of core values for judicial functioning. The judicial cultures project will provide a deeper understanding of specific ways in which core values for judicial functioning are implemented. For each of these two outlooks on the topic, awareness of the limitations of the research is important in order to set realistic goals.
Furthermore, the analysis in this article reveals that researchers should be aware of the difficulties of working with indicators. As noticed elsewhere, awareness is required with regard to underlying assumptions, e.g. economic assumptions in the EU context, and political interests connected to the research. 86 With regard to judicial ethical codes, there could be a risk of scores being used by governments or court managers to put pressure on judges, or by EU institutions to put pressure on member states to push through specific reforms. Another risk is that member states could use scores to downplay deficiencies of their systems. 87 In this regard, it is important to observe that neither of the two surveys discussed in this article aims to rank the examined judicial systems. 88 Also, a precise definition of central terms is essential for generating results which can form the basis of a comparison between national judicial systems. Interdisciplinary perspectives from inter alia the social sciences (sociology, psychology), law, economics, and cultural anthropology can be used to design a sound conceptual and theoretical framework. Interestingly, the two presented surveys display a conceptual difference. The ENCJ's survey considers judicial ethical codes as an element of objective accountability, which relates to the formal arrangements for judiciaries and for judges. By contrast, the judicial cultures project focuses on the perceived influence of judicial ethical codes, which connects with subjective independence and accountability. The combination of the results from these two research projects might, once the second project has been finalised, provide insights into the connections between formal arrangements for judiciaries and the realisation of de facto independence and accountability.
A final preparatory step concerns the choice of research methods. Quantitative empirical research (e.g. surveys) provides numbers, while qualitative empirical research (e.g. interviews) gives insight into the mechanisms behind the numbers. 89 An important practical challenge for empirical research based on surveys or interviews concerns the realisation of a sufficiently high response rate and trustworthiness of responses. Also, researchers should be aware of the risk of a selective non-response bias, which means that only respondents with an interest or experience relating to the topic will participate. However, completed research projects which involved judges demonstrate that these risks can be overcome, e.g. by using a sufficiently large sample of respondents and by checking responses against information from other available sources. 90 Of course, the data from surveys and interviews should be collected, processed, and stored in accordance with applicable norms of academic research ethics. 91 A further concern on the ENCJ's survey design, which aims to serve practical goals, is that subjective indicators are not "actionable". 92 By contrast, a survey based on objective indicators can be replicated and might therefore have more credibility, e.g. for policy advice. In order to remedy this, it could be useful to measure how many times judges have been dismissed rather than to measure the perceptions which judges have of de facto independence. 93 Still, perceptions can serve as a proxy for identifying problems with de facto independence in the examined countries. In the analysis, it should then be emphasised that perceptions are not experience. Also, information from other sources can be consulted to put the results in perspective. 94 It is useful to already consider these possibilities and limitations in the preparatory phase of the research in order to ensure that the intended goals of the research can be achieved.

Conducting research: optimising the quality of the results
In the conduct of empirical research based on a survey a first concern is the validity of the results. For the ENCJ survey, the questionnaire on objective indicators was filled out by a "mixed bag" 95 of organisations, as judicial councils have different compositions, competences, and resources. Moreover, in some countries a judicial council does not exist and for these countries the questionnaire was filled out by other governing bodies, such as ministries of Justice. 96 In order to strengthen the trustworthiness of the research, the ENCJ has validated the results through checks performed by an internal expert group. 97 Some further considerations can be kept in mind with an eye to improving the validity of survey research. The external validation workshop on the ENCJ survey 98 yielded the following suggestions: -Case studies can provide richer information on the situation in different countries. As a means of selection, the method of grouping of indicators (clustering) would make it possible to identify 'red flags' 99 in specific countries. An example of clustering with regard to judicial systems could concern indicators on corruption, the number of disciplinary proceedings against judges, and the length of proceedings. A case study could then focus on countries where these topics in combination seem problematic.
-A longitudinal analysis for each country can provide further insight into the development of the values of judicial independence and accountability.
-The combination of a survey with interviews enables checking the validity of the survey results as well as collecting more indepth information for explaining the results.
-It is important to avoid insularity in the collection of data. This can be realised by asking the same questions to different stakeholders to bring out inconsistencies in answers. Categories of respondents which could be added when studying judicial systems include lawyers, court users, and politicians.
Finally, contextualisation is an important but challenging aspect of comparative research, which also requires attention in research on judicial systems. For the ENCJ survey, critical comments concerned issues relating to the use of indicators, such as the risk of contingent circumstances being moved back in indicators or cultural differences not being acknowledged in the comparative analysis of countries. 100 Remedies can be to relate institutional design of countries to historical developments and to conduct a contextual analysis for each country, in which cultural and other aspects are taken on board. 101 The discussed methodological aspects of research on judicial ethical codes in the framework of the ENCJ survey and the judicial cultures project have been summarised in Table 1.

Conclusion
This article set out to investigate which methodological considerations should be kept in mind when elaborating an empirical research design for studying the influence of judicial ethical codes in EU member states. In the ENCJ's survey on the independence, accountability, and quality of judicial systems in Europe, sub-indicators were included to measure the existence of judicial ethical codes and relating training and supervision. This survey and the critical scholarly discussion about it provide valuable lessons for further comparative and empirical research on the perceptions of judges relating to professional-ethical codes and standards. For the envisaged research in the judicial cultures project, which was presented in this article, the most important lessons can be summarised as follows: 1) to clearly delineate the research aims and central terms which are used in the survey; 2) to avoid insularity by including a representative and diverse group of respondents; 3) to validate and contextualise the results through a multi-method approach and critical debates with scholars and stakeholders. The results of this upcoming research will hopefully contribute to a better understanding of the mille-feuille of judicial ethical codes in the EU and a reflection on a recipe (but not a bake-off!) for the future.