See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/45096979

Research ethics in dissertations: Ethical issues and complexity of reasoning

Article  in  Journal of medical ethics · July 2010

DOI: 10.1136/jme.2009.034561 · Source: PubMed

CITATIONS READS

17 9,183

3 authors:

Sofia Kjellström Sara Nora Ross

Jönköping University Neurotricional Sciences Education Pty Ltd (Queensland)
45 PUBLICATIONS   306 CITATIONS    39 PUBLICATIONS   268 CITATIONS   

SEE PROFILE SEE PROFILE

Bengt Fridlund
Linnaeus University
240 PUBLICATIONS   4,723 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Developmental Complexity Science, political development View project

Financial Incentives and Motivation in Swedish Primary Care View project

All content following this page was uploaded by Sofia Kjellström on 29 July 2015.

The user has requested enhancement of the downloaded file.

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics in dissertations: ethical issues

and complexity of reasoning
S Kjellström, S N Ross and B Fridlund

J Med Ethics 2010 36: 425-430

doi: 10.1136/jme.2009.034561

Updated information and services can be found at:

http://jme.bmj.com/content/36/7/425.full.html

These include:
Supplemental http://jme.bmj.com/content/suppl/2010/07/26/36.7.425.DC1.html
Material
References This article cites 20 articles, 9 of which can be accessed free at:
http://jme.bmj.com/content/36/7/425.full.html#ref-list-1

Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.

Notes

To order reprints of this article go to:

http://jme.bmj.com/cgi/reprintform

To subscribe to Journal of Medical Ethics go to:

http://jme.bmj.com/subscriptions
 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

Research ethics in dissertations: ethical issues and

complexity of reasoning
S Kjellström,1 S N Ross,2,3 B Fridlund4

< Additional data are published ABSTRACT ethics as reflected in the entire research process.
online only. To view these files Background Conducting ethically sound research is Scientific journals increasingly require that ethical
please visit the journal online a fundamental principle of scientific inquiry. Recent considerations are elucidated, but most journals
(http://jme.bmj.com).
1
research has indicated that ethical concerns are severely limit space for elaboration.4 Since studies
Institute of Gerontology, School insufficiently dealt with in dissertations. have questioned the ethical skills of doctoral
of Health Sciences, Jönköping
University, Jönköping, Sweden Purpose To examine which research ethical topics were students, dissertations provide a forum for students
2
Antioch University Midwest, addressed and how these were presented in terms of to expound on ethics and enable an assessment of
Yellow Springs, Ohio, USA complexity of reasoning in Swedish nurses’ dissertations. acquired proficiencies. One purpose of graduate
3
ARINA, Inc., Cincinnati, Ohio, Methods Analyses of ethical content and complexity of school is to train doctoral students in skills neces-
USA
4
Department of Nursing, School
ethical reasoning were performed on 64 Swedish nurses’ sary for future research careers, including more
of Health Sciences, Jönköping PhD dissertations dated 2007. critical thinking and more complex reasoning. The
University, Jönköping, Sweden Results A total of seven ethical topics were identified: quality and depth of the research ethics section is
ethical approval (94% of the dissertations), information essential to examine whether a researcher has
Correspondence to and informed consent (86%), confidentiality (67%), acquired necessary skills to reflect and report on
Sofia Kjellström, Institute of
Gerontology, School of Health
ethical aspects of methods (61%), use of ethical ethics.
Sciences, Jönköping University, principles and regulations (39%), rationale for the study Despite an increasing interest in research ethics,
PO Box 1026, SE-551 11 (20%) and fair participant selection (14%). Four of those surprisingly little is known about the quality of
Jönköping, Sweden; of topics were most frequently addressed: the majority research ethics in dissertations, particularly in
[email protected] of dissertations (72%) included 3e5 issues. While many nursing research. Research on written materials
Received 20 November 2009 ethical concerns, by their nature, involve systematic focuses primarily on research review boards5e9 and
Revised 22 February 2010 concepts or metasystematic principles, ethical reasoning journalsdfor example, ethics guidelines10 and
Accepted 2 March 2010 scored predominantly at lesser levels of complexity: research ethics in articles.4 Research on Turkish
abstract (6% of the dissertations), formal (84%) and nursing dissertations showed deficiencies in
systematic (10%). informing participants and protecting privacy.11 A
Conclusions Research ethics are inadequately covered study on Swedish nurses’ dissertations from 1987
in most dissertations by nurses in Sweden. Important to 2007 showed that an increase in occurrence and
ethical concerns are missing, and the complexity of proportions of reported ethical considerationsand
reasoning on ethical principles, motives and implications that the texts were short, had few references and
is insufficient. This is partly due to traditions and norms covered a narrow range of topics.12 We found no
that discount ethical concerns but is probably also other studies that address the design of the research
a reflection of the ability of PhD students and supervisors ethics section and how different topics were
to handle complexity in general. It is suggested that the combined.
importance of ethical considerations should be The study’s purpose was to examine which
emphasised in graduate and post-graduate studies and research ethical topics were addressed and how these
that individuals with capacity to deal with systematic were presented in terms of complexity of reasoning
and metasystematic concepts are recruited to senior in Swedish nurses’ dissertations approved in 2007.
research positions. The research questions were: Which research ethics
issues are reported? How is the research ethics
section organized around different ethical issues?
How is the information coordinated in terms of the
Research has a potential to encroach on people’s complexity of reasoning that structures the text?
lives, autonomy and integrity. To prevent or miti- What is the relationship between ethical issues and
gate the potential for such effects, the research complexity of reasoning in the text?
community has created ethics codes and regula-
tions, institutionalised ethics review boards and METHODS
formalised ethics requirements in scientific jour- Design and methodological approaches
nals.1e3 However, how do we know whether the The study used a mixed-methods approach to address
formalisations of research ethics actually result in the four research questions.13 We performed a quali-
researchers’ ability to operationalise ethics in the tative content analysis and a quantitative analysis of
ways intended? One way is to analyse how they the hierarchical complexity of ethics-related content.
write about research ethics. The quantification method was the Hierarchical
Including a well-written section about research Complexity Scoring System (HCSS) (Commons, et
ethics in a dissertation is important for several al, unpublished manual), which derives from the
reasons. Compared to protocols written for Model of Hierarchical Complexity, a mathematics-
research ethics committees, this section allows a based, formal general theory applicable to all actions
comparison of the expected and actual research in which information is organised.14 15All reasoning

J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561 425

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

involves organising information. The theory and validated scoring Table 1 Common range of stages of performance in adult tasks’
method enable reliable measures of discrete stages of reasoning hierarchical complexity
complexity.16e20 In accord with Swedish law, ethical approval was Stage Stage name General descriptions of tasks performed
not obtained for this study,21 but ethical principles were used and 8 Concrete Uses simple deductions; time sequences are used to
issues were addressed in ongoing reflective processes. describe actual instances that occur in past or present
time, and deal with specific things, specific incidents or
Data collection events, specific actors, specific actions, specific places.
9 Abstract Uses dualistic stereotypes (eg, good/bad, right/wrong);
The sample consisted of 64 dissertations from Swedish univer- forms abstract variables out of concrete classes of objects;
sities in 2007 (Appendix 1). The primary inclusion criteria were makes and quantifies propositions; uses variable time,
that the dissertation was written by a nurse and that it was place, act, actor, state, type; uses quantifiersd for
a PhD dissertation (4 years of full-time studies). Suitable example, all, none, some and categorical assertionsdfor
example, “We all die”. Attempts to justify an assertion take
dissertations were identified from the Swedish Society of the form of another assertion without specifying a logical
Nursing’s list of self-reported dissertations (n¼65) followed by connection; circular reasoning can appear.
a systematic comparative analysis with the Swedish National 10 Formal Uses empirical or logical evidence; logic is linear, one-
Library (n¼1). One of the self-reported dissertations discussed dimensional (ie, one input variable); solves problems with
one unknown using algebra; forms logical relationships
no research ethics and one was by an unsuccessful doctoral between abstract variables. Uses terms such as if.then,
candidate: they were not included in the sample. Dissertation thus, therefore, and because in justifying assertions.
languages were English (n¼48), Swedish (n¼15) and Norwegian 11 Systematic Constructs multivariate systems and matrices,
(n¼1). Dissertations were retrieved via full-text online access or coordinating more than one variable as input; situates
events and ideas in a larger context, that is, considers
as books from the university library. relationships in contexts; forms systems out of multiple
formal relations.
Data analysis
12 Metasystematic Integrates systems to construct metasystems, including
The dissertations were examined to identify research ethics principles, out of disparate systems; compare systems and
sections, often under the subheadings “Ethical considerations” or perspectives in a systematic way (across multiple
“Ethical approval”. The texts were analysed for the topics domains); reflect on systems, that is, is metalogical, meta-
analytic; name properties of systems (eg, homomorphic,
addressed and how they were reported. An unstructured matrix isomorphic, complete, consistent, commensurable).
of research ethics issues was created and grounded in the data.
Note: From ‘Hierarchical Complexity Scoring System (HCSS) How to Score Anything’ (pp.
The coded texts were further analysed for subcategories through 62; 92e93), by M. L. Commons, P. M. Miller, E. A. Goodheart, and D. Danaher-Gilpin, 2005,
an inductive process. Descriptions of meanings of quantitative Cambridge, MA: Dare Association, Inc. Copyright 1991e2005 by Dare Association, Inc.
and qualitative character, that is manifest and latent content Adapted with permission.

analysis, were sought. The analysis was performed by SK with

BFdwith extensive experience in qualitative methods. parency concern was that several sections included no name of
In hierarchical complexity scoring, such content is “seen the ethics board and/or registration number (n¼13). A minority
through” to examine its underlying structure. The method related the issue of ethics approval to ethical codes, the Helsinki
measures the levels of abstraction and how information is coor- declaration or current national research ethics laws (n¼14) by
dinated. Each section and subsection of a research ethics discus- either stating that studies were performed in accordance with
sion was assessed on stage of hierarchical complexity. The overall ethics regulations (n¼8) or by arguing against the need for an
discussion was scored based on the highest stage of performance ethics approval due to national laws (n¼6).
the text demonstrated. The correlation of content and its
complexity indicated which topics were addressed at different Information and informed consent
stages of complexity. Scoring was performed independently by SK We broadened the traditional informed consent category to
and SR, then discussed to reach consensus. Both authors scored accommodate information-giving processes discussed but not
the English texts, and SK scored the ones in Scandinavian always expressed in terms of informed consent. Most disserta-
languages and discussed with SR. SR is an expert HCSS stage- tions discussed information-giving and informed consent
and-transition scorer while SK is a qualified HCSS scorer of stages (n¼56). A third of these explicitly mentioned the concept of
8 through 11. See table 1 for stage complexity information.22 informed consent (n¼19). A substantial amount of space was
typically used to detail the informing phase of research,
RESULTS including the information’s form (written and/or verbal) (n¼41)
Research ethics issues in dissertations and type. The most often-given information was freedom to
Dissertations contained one to seven research ethics topics: withdraw from the study (n¼33) and a declaration of volun-
approval of research ethics board (94%); information process and tariness (n¼30). Other information included confidentiality
informed consent (86%); confidentiality (67%); ethical aspects of (n¼22), withdrawals’ non-interference with further treatment
methods (61%); use of ethical principles and regulations (39%), (n¼7), the right to not answer questions (n¼4), aim of the study
rationale for the study (20%) and fair participant selection (n¼2), risks and benefits (n¼2) and feedback of results (n¼1).
method (14%; table 2). All but three of the dissertations involved Those responsible for providing information as well as those
direct interaction with study participants; three were register- receiving the information were described. Some informed
based studies. consent discussions included an ethical rationale for the infor-
mation process by referring to principles, codes or laws (n¼16).
Ethics approval
The ethics approval category included descriptions of whether Confidentiality
the dissertation has been vetted by an ethics review board. Items coded in the confidentiality category reported that infor-
Almost all dissertations included a discussion of ethics approval mation was accessible to only authorised persons. Confidenti-
(n¼60), and a majority stated they had been approved by ality procedures were succinctly reported (n¼43). Besides
a research ethics review board (n¼55). A quality and trans- describing confidentiality as something that participants were

426 J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

Table 2 Design of research ethics sections in Swedish nurses’ dissertations

Number of topics
Hierarchical complexity
on research ethics Dissertations in
in the dissertation the category Topics on research ethics of reasoning
Ethical Informed Ethical aspects of Principles, Fair participation
approval consent methods Confidentiality regulations Rationale selection Score 9 Score 10 Score 11
1 4 4 2 2 0
Sum 4 4 2 2 0

2 4 4 4 1 3 0
2 2 2 2 0
2 2 2 1 1 0
1 1 1 1 0
Sum 9 8 4 3 3 2 8 0

3 5 5 5 5 0 5 0
5 5 5 5 0 5 0
1 1 1 1 0 0 1
1 1 1 1 0 1 0
1 1 1 1 0 1 0
1 1 1 1 0 0 1
Sum 14 11 14 6 7 2 1 1 0 12 2

4 8 8 8 8 8 0 7 1
4 4 4 4 4 0 4 0
3 3 3 3 3 0 2 1
1 1 1 1 1 0 1 0
1 1 1 1 1 0 1 0
Sum 17 17 17 12 12 5 4 0 15 2

5 9 9 9 9 9 9 0 8 1
2 2 2 2 2 2 0 2 0
1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 0 1 0
1 1 1 1 1 1 0 1 0
1 1 1 1 1 1 0 1 0
Sum 15 15 14 11 15 12 5 3 0 13 2

6 3 3 3 3 3 3 3 0 3 0
1 1 1 1 1 1 1 0 1 0
1 1 1 1 1 1 1 0 1 0
Sum 5 5 5 5 4 5 2 4 0 5 0

7 1 1 1 1 1 1 1 1 0 1 0
Sum 1 1 1 1 1 1 1 1 0 0 0

Total 64 60 55 39 43 25 13 9 4 54 6
Per cent 100 94 86 61 67 39 20 14 6 84 10

guaranteed and informed about, some researchers identified how Strategies to impede negative consequences were depicted
confidentiality had been handled: data were safely stored (n¼20): adopt a sensitive attitude, adapt to the physical and
protecting participant’s identity (n¼12); data were analysed and mental status of the interviewee, reduce questions, provide time
reported without identifying participants (n¼19) and partici- to reflect on the interview and arrange for a contact person.
pants in focus group interviews were counselled in ways to Sometimes, statements about how the participants seemed to
promote freedom of expression and confidentiality (n¼2). enjoy the interview experience were included (n¼14). A few
sections described problems that appeared during the research
Ethical aspect of the methods interview (n¼14)dfor example, interviewees who cried or did
The category for ethical aspect of the methods included the not answer all questions. The most comprehensive sections
research ethics issues in collecting data, except for questions covered all these issues, but the most common strategy was to
regarding informing participants. Ethical aspects of study mention the potential laboriousness of the interview yet argue
methods were comprised of descriptions of interviews and that participants benefited from practical solutions that were
questionnaires (n¼37). Explanations of why interviews were provided in the interview situation or by claiming that research
ethically problematic were done by referring to principles or participants appreciated the opportunity to tell their stories. The
risks of harm (n¼17). The negative aspects stated (n¼24) were reported ethical problems with questionnaires were primarily
physical and psychological with an emphasis on emotional. the tedium of answering questions and how researchers adjusted

J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561 427

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

the number of requests for completion out of respect and logic or evidence to justify the assertiondfor example, by
concern for participants’ possible fatigue. providing a logical explanationdfor example, using such terms
as because, in order to, since, if, then, therefore. Descriptions of
Use of ethical principles and regulations hypothetical or alternative options in the future were sometimes
Discussions that included the usage of principles and ethical included. The logic was linear. Such linear logic took the form of
regulations like laws and research ethics codes were coded to the ifethen constructions or chains of logic. Some used principles as
category of ethical principles and regulations. This category was logical reasons for actions.
analytically different from others because it revealed how ethics Systematic stage performances were characterised by the
were applied in the research sections. Explicit report of laws, ability to coordinate at least two logical relations into a system
ethics codes and principles occurred in fewer than half of the (table 3); in other words, they demonstrated reasoning about
dissertations (n¼25). Principles were employed but performed in complex causation and ability to understand a system of logical
qualitatively different ways (n¼17). The simplest form was to relationships. For example, one researcher described procedures
state that the study had been performed in accordance with for finding the “right people” by invoking a multivariate system
a research ethics declaration, code or rules outlined in a research that required the coordination of multiple variables. Systemic
ethics book. The most elaborate ones integrated the principles stage performances were characterised by more fluid reasoning
and described how they were used as compasses for research than the linear, logical performances.
procedures (n¼8).
Comparing content and complexity
Rationale for the study Few dissertations demonstrated abstract reasoning and system-
To provide an ethical rationale for the study means to justify atic reasoning, four and six, respectively, but showed interesting
why the study is important in a wider perspective. Thirteen patterns. The texts with abstract stage reasoning reported either
dissertations featured an ethical rationale for the study, and one or two topics. All four mentioned approval; information and
when included, it was framed in terms of risks and benefits. The methodological issues were raised by only two. Texts with
need for new and valuable knowledge that could potentially systematic reasoning introduced three to five ethical issues. Half
improve conditions for other people weighed heavier than the of them discussed principles (as compared to merely citing
extra demand and little direct gain that the research subjects a principle as the reason for an action), and the other three
gained from participating. Some reported that the value of reported the rationale for the study, indicating that the topic and
pursuing the research outweighed the disadvantages but entailed study could perhaps be viewed in a wider context. Among the
the necessity of protecting the autonomy of the research majority of texts demonstrating formal reasoning, the topics
participants. varied from one to seven, meaning at least formal reasoning was
needed to explain all conceivable aspects. Formal reasoning is
Fair participant selection required to report such tasks as fair selection of participants,
Fair selection of participants signifies reflections on a justified rationale for the study and principles, ethics codes and laws.
choice of participants. The reason to include vulnerable groups
and groups that previously has been excluded from research was DISCUSSION
sometimes given (n¼9). A few sections justified the choice of Our study demonstrates that research ethics are insufficiently
participants (n¼8). The importance of including important and reported and inadequately described in many nursing disserta-
vulnerable groups so their voices would be heard was the main tions. Few ethical topics are considered, and they are not
reason reported. discussed in a thorough way. While most note official approval
and describe informed consent issues, other issues like the
Design of the research ethics section rationale for the study and how the participants were selected
The topics of the research ethics sections are outlined in table 2. are infrequently reported. The level of complexity of reasoning
Most frequent was to report four ethics issues (n¼16), followed was inadequate in most dissertations. The majority of the
by three (n¼14) or five issues (n¼ 14). The majority (72%) dissertations used formal reasoning, although by their nature,
included 3e5 issues. Four sections stated one topic and only one the ethical issues introduced in them require more complex
dissertation section reported seven issues. The most common reasoning to be satisfactorily addressed.
composition of a section about research ethics discussed five A methodological strength of our study is its inclusion of
topics: the approval from a research review board, information a large number of dissertations, which are likely representative
and informed consent, ethical aspects of the methods, confi- of dissertations by Swedish nurses. A major advantage of our
dentiality and principles. method is that the analytical approach permits assessments and
comparisons of the coverage of ethical issues and the complexity
Complexity of reasoning of reasoning.
The analysed texts demonstrated three stages of performance as A methodological shortcoming is that the analysis was
measured by hierarchical complexity: abstract (n¼4), formal primarily focused on the section denoted “Ethical consider-
(n¼54) and systematic (n¼6). ations/approval”, thus some ethics topics and reasoning might
Abstract stage text performances consisted of declarative have passed undetected if they were treated in other parts of the
statements (table 3). Unsupported categorical assertions were dissertation. The analysis is thus limited to what the authors
made and justified by invoking another assertion. Generalisa- define as belonging to ethics sections. Our analysis identified the
tions were created by quantifying people and events. Often-used most complex stage of reasoning as a criterion for analysis
quantifications in the sample were “all participants” and “all because ethical considerations are complex matters. A more
studies”. Research ethics sections included mainly generalisa- extensive analysis could have also analysed the entire low to
tions about actions that had been performed. high range of reasoning demonstrated in each ethics section. An
Reasoning at the formal stage of performance used empirical or implication of the language analyses is that we do not know
logical evidence (table 3). Assertions were supported by explicit which and how the ethical issues were applied in reality. Some

428 J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

Table 3 Representative examples of reasoning in research ethics at three stages of complexity

Topic Abstract reasoning Formal reasoning Systematic reasoning
Who informed As a second step, they were contacted by A coordination nurse at first approached By getting help from a gate-opener in
a nurse over the phone and asked the parents in order to let them decide finding suitable participants in the
whether or not they were willing to about participation, and to ask whether research studies the risk of exposing
participate. (Dissertation 6) the researcher could take contact with sensitive and too vulnerable individuals
them. (Dissertation 4) was eliminated, which upholds the
principle of non-malfeasance (Beauchamp
and Childress, 1994). (Dissertation 53)
Information All participant were given written (I, II, Supported by the principle of beneficence, No example.
IV), or written and oral information (III) information about the research was
about the study and that participation was therefore given verbally and in writing,
voluntary. (Dissertation 58) including information about voluntary
participation. (Dissertation 53)
Risk with interviews No example. Since the patients were interviewed about Despite the risk that interview questions
their experiences that could have strong might cause feelings of uneasiness, there
emotional meaning and could be was deemed to be a greater potential for
distressful to talk about, they were offered beneficence that outweighed the risks by
an opportunity to contact the interviewer providing participants with the
after the interviews to discuss any opportunity to tell their stories in an open
subsequent reflections. (Dissertation 18) and encouraging way. (Dissertation 13)

issues could have been omitted from the dissertation text even more elaborated use of ethics principles. Very few used systematic
though the issue was dealt with in practice and vice versa. The reasoning, and none used metasystematic, which would be pref-
consistency between writing about ethics and ethical behaviour erable because several of the research ethics concepts are meta-
in the fielddfor example, in contact with research subjects and systematic stage principles. For example, informed consent is
patients, should be investigated in future studies. a metasystematic stage concept because it coordinates the system
The first main finding is the incompleteness of the elaboration of informing a research subject and the system of obtaining
of topics and details in several dissertations, which is consistent consent from the person.24 This means that metasystematic
with several studies in the domain of research ethics. A previous reasoning is needed for a full understanding and use of these
study showed a high level of errors in research ethics committee concepts.
letters; that is, procedural violations, missing information, slip- What are possible explanations for the low levels of reasoning
ups and discrepancies.8 Earlier research on Swedish nurses’ on research ethics? One possibility is that ethical issues are dealt
dissertations demonstrate the questionable quality due to short with at a sufficiently high level of complexity in practice, whereas
length, few references and a narrow range of topics.12 the text of the dissertation merely reflects a research tradition
In our study, few topics were addressed. Emanuel et al argued that discounts the importance of performing and explaining
for seven requirements to be considered and met in the conduct ethical reasoning. Disciplinary norms for terse writing styles are
of ethical research: scientific value, validity, fair subject selection, presumably promoted by supervisors and department guidelines.
favourable riskebenefit ratio, independent review, informed For example, nurses’ dissertations in social science use more
consent and respect for potential and enrolled subjects.23 references to methods, ethics and philosophy of science than
Applied to our findings, some requirements may be treated in dissertation in the medical science tradition.23 In addition, poor
other parts of a dissertation, but several dissertations leave out writing may occur because researchers mimic previous disserta-
topics that are necessary for judging their ethical quality. tions or regard ethical considerations as bureaucratic hurdles rather
Informing potential participants and pursuing informed than moral requirements to protect participants. The supervisor
consent was reported in almost 90% of the dissertations’ ethics role is an important factor since they sometimes acknowledge
sections. This frequency is higher than that reported in a study of a considerable lack of knowledge about research ethics.25 Another
Turkish nurses’ dissertations where subjects were not informed conceivable explanation is that the level of ethical reasoning
about the study (72.7%) and the researchers had not obtained corresponds rather accurately to the level of complexity the
permission from the subjects (73.6%).11 doctoral students and their supervisors use to handle complex
The second main finding is the insufficient level of complexity issues in general. In other words, they are arguing on ethical issues
of reasoning, with which research ethics are handled. Findings at their highest complexity level. In that case, the scientists’ (PhD
from a discourse analysis of research ethics committee letters students’ and supervisors’) ability to discuss at more complex
showed that there was “the lack of formal reasoning” (p 258) and levels must be improved for ethical issues to be sufficiently
ethical argumentsdfor example, informed consent are described managed in the future. All these possibilities suggest further
as procedural norms rather than an ethics principle possible to research is needed to account for our findings, since ethics have
dispute.9 This is consistent with our findings, because a signifi- long been an important part of nurses’ education and occupation.
cant number invoked research ethics principles to justify proce- There are several implications of insufficient ethical reasoning.
dures taken, rather than to use principles to support ethical Integrity of the research subjects and patients are at risk, and
arguments for and against certain procedures. However, our patients, if they participate, may be informed without under-
findings also showed that the great majority used at least some standing the implications. From the perspective of the readers of
formal reasoning, as measured by hierarchical complexity. the scientific literature, it is impossible to assess how and why
Unfortunately, formal reasoning is necessary but not suffi- the authors dealt with various ethical issues. A crucial implica-
cient for adequacy in ethical matters. The analysis showed that tion is the consequences of selection of research questions,
formal reasoning and systematic reasoning were needed to methods and participants/sample. Scientists performing at
elaborate on topics, and the comparison of complexity reasoning abstract or formal stages are less likely to integrate relevant
and content indicated that higher levels of reasoning involved ethical aspects into their research aims than scientists at higher

J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561 429

 Downloaded from jme.bmj.com on August 9, 2010 - Published by group.bmj.com

Research ethics

complexity levels. This is because such integration, by its nature, 3. Fischer BA. A summary of important documents in the field of research ethics.
is multivariate at minimum. They will differ quite dramatically Schizophr Bull 2006;32:69e80.
4. Finlay KA, Fernandez CV. Failure to report and provide commentary on research
in the way they understand principles as principles, “risks” and ethics board approval and informed consent in medical journals. J Med Ethics
“benefits”, rationale of the investigation, etc. Researchers with 2008;34:761e4.
systemic or metasystematic stage reasoning are able to ask more 5. Dixon-Woods M, Angell EL. Research involving adults who lack capacity: how have
research ethics committees interpreted the requirements? J Med Ethics
complex questions, juggle ethics, research questions, and 2009;35:377e81.
methods and design more complex research projects.26 6. Angell EL, Jackson CJ, Ashcroft RE, et al. Is ‘inconsistency’ in research ethics
Our conclusion is that if the established praxis to include committee decision-making really a problem? An empirical investigation and
reflection. Clin Ethics 2007;2:92e9.
discussion of research ethics in Swedish nurses’ dissertations is 7. Angell EL, Bryman A, Ashcroft RE, et al. An analysis of decision letters by research
going to be valuable, and if its purpose is to indicate that the ethics committees: the ethics/scientific quality boundary examined. Qual Saf Health
research complied with expected ethics, then the reporting must Care 2008;17:131e6.
exhibit a certain quality, comprehensiveness and sufficiently 8. Angell E, Dixon-Woods M. Do research ethics committees identify process errors in
applications for ethical approval? J Med Ethics 2009;35:130e2.
significant treatment of ethics. Our study illustrates that factors 9. O’Reilly M, Dixon-Woods M, Angell E, et al. Doing accountability: a discourse
that improve the quality include: appropriately thorough analysis of research ethics committee letters. Sociol Health Illn 2009;
consideration of several ethical issues while avoiding minutiae; 31:246e61.
10. Rowan-Legg A, Weijer C, Fernandez CV. A comparison of journal instructions
use of ethical principles in appropriate contexts to justify choices regarding institutional review board approval and conflict-of-interest disclosure
and reasons to support actions taken and use of at least formal between 1995e2005 J Med Ethics 2009;35:74e8.
and systematic reasoning. In addition, we would like to see more 11. Ulusoy MF, Ucar H. An ethical insight into nursing research in Turkey. Nurs Ethics
reflection and a critical stance to what has been done in the 2000;7:285e95.
12. Kjellström S, Fridlund B. Status and trends of research ethics in Swedish nurses’
dissertation work. dissertations. J Nurs Ethics 2010;17:382e3.
In order to accomplish the intent of reporting research ethics, 13. Creswell JW. Research design: qualitative, quantitative, and mixed methods
several improvements are needed. The most straightforward approaches. Thousand Oaks (CA): Sage, 2009.
14. Commons ML, Pekker A. Presenting the formal theory of hierarchical complexity.
solution is to enhance the research ethics teaching in graduate World Futures 2008;64:375e82.
education. Students must learn how to perform ethically sound 15. Commons ML, Goodheart EA, Pekker A, et al. Using Rasch scaled stage scores to
research from the first steps of planning and performing to writing validate orders of hierarchical complexity of balance beam task sequences. In: Smith
JEV, Smith RM, eds. Rasch measurement: advanced and specialized applications.
up the results and their potential and ability to report and reflect Maple Grove (MN): JAM Press, 2007;121e47.
on ethical aspects of the research process must be enhanced. A 16. Dawson TL. A comparision of three developmental stage scoring systems. J Appl
more profound resolution is to emphasise metasystematic Meas 2002;3:146e89.
thinking in post-graduate studies and recruit senior researcher and 17. Dawson TL. A stage is a stage is a stage: a direct somparison of two scoring
systems. J Genet Psychol 2003;164:335e64.
post-graduate students who already have developed a systematic 18. Dawson TL, Xie Y, Wilson M. Domain-general and domain-specific developmental
or metasystematic way of reasoning. This longer-term solution assessments: do they measure the same thing? Cogn Dev 2003;18:61e78.
will also constitute the foundation for further development of 19. Dawson TL. The meaning and measurement of conceptual development in
adulthood. In: Hoare CH, ed. Handbook of adult development and learning. Oxford:
complexity in handling ethics issues in the future. Oxford University Press, 2006:433e54.
20. Dawson TL. Assessing intellectual development: three approaches, one sequence.
Acknowledgements We would like to thank professor Per Sjölander for valuable J Adult Dev 2004;11:71e85.
comments on the discussion. 21. Swedish law. Act on Ethical Review of Research Involving Humans, 460. 2003.
22. Commons ML. Introduction to the model of hierarchical complexity and its relationship
Competing interests None. to postformal action. World Futures, The Journal of General Evolution 2008;64:305e20.
Provenance and peer review Not commissioned; externally peer reviewed. 23. Heyman I. Gånge hatt till: omvårdnadsforskningens framväxt i
Sverigedsjuksköterskors avhandlingar (The emergence of nursing research in
Swedendnurses’ dissertations). Göteborg: Daidalos; 1995.
24. Commons ML, Rodriguez JA, Adams KM, et al. Informed consent: do you know it
REFERENCES when you see it? Psychiatr Ann 2006;36:430e5.
1. Emanuel EJ, Wendler D, Grady C. What makes clinical research ethical. JAMA 25. Szirony TA, Price JH, Wolfe E, et al. Perceptions of nursing faculty regarding ethical
2000;283:2701e10. issues in nursing research. J Nurs Educ 2004;43:270e9.
2. Dixon-Woods M, Ashcroft RE. Regulation and the social license for medical 26. Commons ML, Richards FA. Four postformal stages. In: Demick J, ed. Handbook of
research. Med Health Care Philos 2008;11:381e91. adult development. New York: Plenum, 2002:199e219.

430 J Med Ethics 2010;36:425e430. doi:10.1136/jme.2009.034561

View publication stats