qualitative content analysis

Nurse Education Today (2004) 24, 105–112

NurseEducationToday

intl.elsevierhealth.com/journals/nedt

Qualitative content analysis in nursing research:concepts, procedures and measures to achievetrustworthiness

U.H. Graneheim*, B. Lundman

Department of Nursing, Ume�a University, Ume�a 90187, Sweden

Accepted 8 October 2003

Summary Qualitative content analysis as described in published literature showsconflicting opinions and unsolved issues regarding meaning and use of concepts,procedures and interpretation. This paper provides an overview of importantconcepts (manifest and latent content, unit of analysis, meaning unit, condensation,abstraction, content area, code, category and theme) related to qualitative contentanalysis; illustrates the use of concepts related to the research procedure; andproposes measures to achieve trustworthiness (credibility, dependability and trans-ferability) throughout the steps of the research procedure. Interpretation inqualitative content analysis is discussed in light of Watzlawick et al.’s [Pragmaticsof Human Communication. A Study of Interactional Patterns, Pathologies andParadoxes. W.W. Norton & Company, New York, London] theory of communication.

�c 2003 Elsevier Ltd. All rights reserved.

KEYWORDSCredibility;Dependability;Latent content;Manifest content;Nursing;Qualitative content

analysis;Transferability;Trustworthiness

Introduction

Initially content analysis dealt with ‘the objective,systematic and quantitative description of themanifest content of communication’ (Berelson,1952, p. 18) but, over time, it has expanded to alsoinclude interpretations of latent content. Manyauthors, from a variety of research traditions, haveaddressed content analysis (for example, Berelson,1952; Krippendorff, 1980; Findahl and H€oijer,1981; Woods and Catanzaro, 1988; Downe-Wam-boldt, 1992; Burnard, 1991, 1996; Polit and Hun-gler, 1999). The first descriptions date from the1950s and are predominately quantitative. Cur-

* Corresponding author. Tel.: +46-90-786-9258; fax: +46-90-786-9169.

E-mail address: [email protected] (U.H.Graneheim).

0260-6917/$ - see front matter �c 2003 Elsevier Ltd. All rights reserdoi:10.1016/j.nedt.2003.10.001

rently, two principal uses of content analysis areevident. One is a quantitative approach often usedin, for example, media research, and the other is aqualitative approach often used in, for example,nursing research and education. Qualitative con-tent analysis in nursing research and education hasbeen applied to a variety of data and to variousdepths of interpretation (for example, O’Brienet al., 1997; Latter et al., 2000; Berg and WelanderHansson, 2000; S€oderberg and Lundman, 2001).

A review of literature based on common data-bases (Cinahl, Medline and Sociological Abstracts)as well as references from articles and books showsdifferent opinions and unsolved issues regardingmeaning and use of concepts, procedures and in-terpretation in qualitative content analysis. Thediversities can be understood partly from a histor-ical point of view and partly from various beliefs ofthe nature of reality among researchers.

ved.

mail to: [email protected]

106 U.H. Graneheim, B. Lundman

An assumption underlying our paper is that re-ality can be interpreted in various ways and theunderstanding is dependent on subjective inter-pretation. Qualitative research, based on datafrom narratives and observations, requires under-standing and co-operation between the researcherand the participants, such that texts based on in-terviews and observations are mutual, contextualand value bound (Lincoln and Guba, 1985; Mishler,1986). Thus, our presumption is that a text alwaysinvolves multiple meanings and there is alwayssome degree of interpretation when approaching atext. This is an essential issue when discussingtrustworthiness of findings in qualitative contentanalysis.

Another issue is that concepts within the quan-titative research tradition still predominate whendescribing qualitative content analysis (for exam-ple, Krippendorff, 1980; Burnard, 1991; Downe-Wamboldt, 1992), especially the use of conceptsdescribing trustworthiness. This causes confusionand paradigmatic uncertainty among authors andreaders of scientific papers.

The purpose of this paper was threefold: first, toprovide an overview of concepts of importancerelated to qualitative content analysis in nursingresearch; second, to illustrate the use of conceptsrelated to the research procedure; and third, toaddress measures to achieve trustworthiness.

Overview of concepts

The following provides an overview of conceptsrelated to qualitative content analysis and is to beseen as a contribution to a debate rather than anendeavour to find consensus. First, we presentvarious uses of concepts found in the literature,and then we give reasons for our stance. The con-cepts are manifest and latent content, unit ofanalysis, meaning unit, condensing, abstracting,content area, code, category and theme.

A basic issue when performing qualitative con-tent analysis is to decide whether the analysisshould focus on manifest or latent content. Analysisof what the text says deals with the content aspectand describes the visible, obvious components, re-ferred to as the manifest content (Downe-Wam-boldt, 1992; Kondracki et al., 2002). In contrast,analysis of what the text talks about deals with therelationship aspect and involves an interpretationof the underlying meaning of the text, referred toas the latent content (Downe-Wamboldt, 1992;Kondracki et al., 2002). Both manifest and latentcontent deal with interpretation but the interpre-tations vary in depth and level of abstraction.

One of the most basic decisions when usingcontent analysis is selecting the unit of analysis. Inthe literature, unit of analysis refers to a greatvariety of objects of study, for example, a person,a program, an organisation, a classroom or a clinic(Mertens, 1998), or a community, state or nation(Patton, 1987). Other authors have considered theunit of analysis as interviews or diaries in theirentity, and the amount of space allocated to atopic or an interaction under study (Downe-Wam-boldt, 1992). Parts of the text that are abstractedand coded (Weber, 1990), or every word or phrasewritten in the transcript (Feeley and Gottlieb,1998), have also been considered as units of anal-ysis. We suggest that the most suitable unit ofanalysis is whole interviews or observational pro-tocols that are large enough to be considered awhole and small enough to be possible to keep inmind as a context for the meaning unit, during theanalysis process.

A meaning unit, that is, the constellation ofwords or statements that relate to the same cen-tral meaning, has been referred to as a contentunit or coding unit (Baxter, 1991), an idea unit(Kovach, 1991), a textual unit (Krippendorff,1980), a keyword and phrase (Lichstein and Young,1996), a unit of analysis (Downe-Wamboldt, 1992),and a theme (Polit and Hungler, 1991). We considera meaning unit as words, sentences or paragraphscontaining aspects related to each other throughtheir content and context.

In the literature, shortening the text includesthe concepts of reduction (Findahl and H€oijer,1981), distillation (Cavanagh, 1997) and conden-sation (Coffey and Atkinson, 1996). Reduction re-fers to decreasing the size, but it indicates nothingabout the quality of what remains. Distillationdeals with the abstracted quality of a text, whichwe see as a further step in the analysis process. Weprefer condensation, as it refers to a process ofshortening while still preserving the core.

The process whereby condensed text is ab-stracted has been called aggregation (Barrosso,1997) and ‘grouping together under higher orderheadings’ (Burnard, 1991, p. 462). We suggest ab-straction, since it emphasises descriptions and in-terpretations on a higher logical level. Examples ofabstraction include the creations of codes, cate-gories and themes on varying levels.

Parts of a text dealing with a specific issue havebeen referred to as a domain or rough structure(Patton, 1990), a cluster (Barrosso, 1997) and acontent area (Baxter, 1991). We prefer contentarea since it sheds light on a specific explicit areaof content identified with little interpretation. Acontent area can be parts of the text based on

Qualitative content analysis in nursing research 107

theoretical assumptions from the literature, orparts of the text that address a specific topic in aninterview or observation guide.

The label of a meaning unit has been referred toas a code. There seems to be agreement in theliterature about the use and the meaning of a code.According to Coffey and Atkinson (1996, p. 32)‘codes are tools to think with’ and ‘heuristic de-vices’ since labelling a condensed meaning unitwith a code allows the data to be thought about innew and different ways. A code can be assigned to,for example, discrete objects, events and otherphenomena, and should be understood in relationto the context.

Creating categories is the core feature of qual-itative content analysis. A category is a group ofcontent that shares a commonality (Krippendorff,1980). Patton (1987) describes categories as in-ternally homogeneous and externally heteroge-neous. Krippendorff (1980) emphasises thatcategories must be exhaustive and mutually ex-clusive. This means that no data related to thepurpose should be excluded due to lack of a suit-able category. Furthermore, no data should fallbetween two categories or fit into more than onecategory. However, owing to the intertwined na-ture of human experiences, it is not always possibleto create mutually exclusive categories when atext deals with experiences. A category answersthe question ‘What?’ (Krippendorff, 1980) and canbe identified as a thread throughout the codes. Aswe see it, a category refers mainly to a descriptivelevel of content and can thus be seen as an ex-pression of the manifest content of the text. Acategory often includes a number of sub-categoriesor sub-subcategories at varying levels of abstrac-tion. The sub-categories can be sorted and ab-stracted into a category or a category can bedivided into sub-categories.

The concept of theme has multiple meaningsand creating themes is a way to link the underlyingmeanings together in categories. Polit and Hungler(1999) describe a theme as a recurring regularitydeveloped within categories or cutting across cat-egories. Baxter (1991) defines themes as threads of

Meaning unit Cond

there is a curious feeling in the head in

some way, empty in some way

c

em

it is more unpredictable so to say, you

can never be sure about anything

An

Figure 1 Examples of meaning units, c

meaning that recur in domain after domain. Theconcept of theme is also used in literature in otherqualitative methods. van Manen (1990, p. 87)considers a theme to ‘describe an aspect of thestructure of experience’ and emphasises that atheme can not be an object or a thing. A themeanswers the question ‘How?’ We consider a themeto be a thread of an underlying meaning through,condensed meaning units, codes or categories, onan interpretative level. A theme can be seen as anexpression of the latent content of the text. Sinceall data have multiple meanings (Krippendorff,1980; Downe-Wamboldt, 1992), themes are notnecessarily mutually exclusive. A condensedmeaning unit, a code or a category can fit intomore than one theme. A theme can be constructedby sub-themes or divided into sub-themes.

Illustrations of the use of concepts

In the following we illustrate the use of conceptsand analysis procedures for two texts based on in-terviews and observations respectively. One ratio-nale behind giving two examples is to show variousways to develop themes. The processes of analysisare described and shown in Figs. 1–3. Even if thesedescriptions point to a linear process, it is impor-tant to bear in mind that the process of analysisinvolves a back and forth movement between thewhole and parts of the text.

Qualitative content analysis of aninterview text

The unit of analysis in this example is interviewtext about experiences of having hypoglycaemia.The context consists of a larger study aimed atdescribing coping strategies related to the every-day strains of living with diabetes (Lundman andNorberg, 1993). Twenty adults with Type 1-diabe-tes, aged 25–59 years, participated in the study.Interviews were performed addressing various as-pects of living with Type 1-diabetes. The interview

ensed meaning unit Code

urious feeling of

ptiness in the head

emptiness in the

head

unpredictable and

unsure situation

uncertainty

ondensed meaning units and codes.

Theme LACK OF CONTROL AND STRUGGLE FOR REGAINING CONTROL

Category SENSATIONS ACTIONS COGNITIONS

Sub-category

Unfamiliar bodily

sensations

Familiar butunexpected

bodily sensations

Terrible feelings Unfamiliaractions

Familiar butunexpected

actions

Thinking difficulties

Codes

Double visionEmptiness in the headEmptiness in the stomachStiffness in the tongueUnsteadinessWeightiness in the legs

Enormous hungerPalpitations of the heart TremorSweating

Frightened to deathLoss of controlUncertainty Confusion

FumblingDisconnected speech Disruptive behaviour

Urge to eatVerbal expressions of anger “Attack” on the refrigerator

Inability to think clearlyUncertainty of what to sayUnawareness of surroundings

Figure 2 Examples of codes, sub-categories, categories and a theme from content analysis of narratives abouthypoglycaemia.

Meaning unit Condensed meaning unitDescription close to the text

Condensed meaning unitInterpretation of the underlying meaning

Sub-theme Theme

She kicks about and hits the care provider when she is putting shampoo to her hair. // She tries to push the care providers away.

Using physical violence when being undressed and washed.

Fighting to defend her body zone against intrusion.

When the care providers are in her room she closes the door from the outside so the care providers are locked up in her room and she stays outside in the corridor.

Closing the door between herself and the care providers.

Marking a boundary against others.

Fighting to protect her personal space

She comes out to the corridor. She wears T-shirt, plastic pants and diapers and she has faeces all over her body. She walks into another resident’s room and locks the door. // The care provider goes to see what she is doing and it appears that she has laid down in his bed.

Appearing undressed and “dirty” in commonly used areas and in other residents’ rooms and beds.

Crossing fellow residents’ physical space.

She goes into the ward office and starts to mess about among the staff’s documents.

Causing a mes s in the ward office.

Crossing the care providers’ physical space.

Invading the physical space of others

The care provider knocks on her door, waits for an answer.

Knocks on the door and waits for an answer.

Asking permission and waiting for an answer before entering her room.

Paying respect to her physical space

The care providers permit her to rise and rummage about, she is allowed to move around while they are looking after her. // She is wandering around in the bathroom during the showering.

Permitting her to rise, rummage about, move around and wander during the morning toilet.

Allowing a certain amount of freedom of movement during the morning toilet.

Paying respect to her personal space

She sits in a chair in her room restrained by a belt. // The care providers put her into a shower chair and restrain her with a belt, which is tied to the back of the chair.

Using physical restraints.

The care provider sits on her bed and leans over her.

Sits on her bed and leans over her.

Coming too close.

Care providers ask: “Shall we go to the toilet?” “Shall we take a shower?”

Addressing her as we instead of you.

The care provider is talking with others about her rash and itch.

Discussing private matters over her head.

Treating private matters as common matters.

Invading her personal space

Interaction as a process of respecting and invading each other’s privacy

Figure 3 Examples of meaning units, condensed meaning units, sub-themes and themes from content analysis ofobservations about interaction between a woman with dementia and her care providers.


text was sorted into seven content areas: experi-ences related to the onset of the disease; man-agement of the disease in daily living; experiencesrelated to hypoglycaemia; experiences related tohyperglycaemia; self-monitoring of blood glucose;and ideas about complications and the future. Ex-periences related to hypoglycaemia were evokedby asking: ‘Please tell me about your experiencesof having hypoglycaemia.’

The interviews were read through several timesto obtain a sense of the whole. Then the text about

the participants’ experiences of having hypo-glycaemia was extracted and brought together intoone text, which constituted the unit of analysis.The text was divided into meaning units that werecondensed. The condensed meaning units wereabstracted and labelled with a code. Examples ofmeaning units, condensed meaning units and codesare shown in Fig. 1. The whole context was con-sidered when condensing and labelling meaningunits with codes. The various codes were comparedbased on differences and similarities and sorted


into six sub-categories and three categories, whichconstitute the manifest content. The tentativecategories were discussed by two researchers andrevised. What differed between the two research-ers was their judgement about what comprisedfamiliar and unfamiliar sensations and actions. Aprocess of reflection and discussion resulted inagreement about how to sort the codes. Finally,the underlying meaning, that is, the latent con-tent, of the categories, was formulated into atheme. Examples of codes, sub-categories, cate-gories and a theme are given in Fig. 2.

Qualitative content analysis of a textbased on observations

The unit of analysis in this example is text based on14 observational notes and six reflective dialogues.The context was a study aiming to illuminate howone woman with dementia and ‘behavioural dis-turbances’ acted in relation to her care providers,and how the care providers acted in relation to her(Graneheim et al., 2001). The study was performedat a residential home for people with dementia andso called ‘behavioural disturbances’. The careproviders were asked to select a person whose‘behavioural disturbances’ caused severe difficul-ties in daily care. Two observers participated on sixoccasions during morning toilet and breakfast. Oneobserver was familiar to the setting and the par-ticipants and represented an insider perspective.The other observer was unfamiliar with theseconditions and represented an outsider perspec-tive. The participant observations focused on theinteraction going on between the woman with de-mentia and her care providers. To further illumi-nate various aspects of the ongoing interaction, areflective dialogue between the observers and thecare providers followed each observation occasion.The observational notes and reflective dialogueswere tape-recorded and transcribed verbatim.

The text was read through several times to ob-tain a sense of the whole. Six observational notes,one from each occasion, were divided into meaningunits. Considering the context, the meaning unitswere condensed into a description close to thetext, the manifest content, and, where possible,into an interpretation of the underlying meaning,the latent content. Since parts of the text weremuch more concentrated than an interview text,further condensation was difficult. The condensedmeaning units were seen as a whole and abstractedinto sub-themes. Examples of meaning units, con-densed meaning units, sub-themes and theme areshown in Fig. 3. Sub-themes were threads of

meaning running through the condensed text. Thesub-themes were presented to the care providersand revised with consideration to their opinion.The remaining eight observational notes wereanalysed. A co-researcher read one-third of theobservational notes and the thematisation. A pro-cess of reflection and discussion resulted inagreement on a set of sub-themes. Lastly, reflec-tion on the sub-themes and a review of literaturerelated to the sub-themes provided phenomenathat seemed to serve as relevant headings to unifythe sub-themes into themes. To reveal meaningunits that rejected interpretations of the observa-tional text the reflective dialogues were analysedand nothing that contradicted the themes could befound.

Measures for achieving trustworthiness

Research findings should be as trustworthy as pos-sible and every research study must be evaluated inrelation to the procedures used to generate thefindings. The use of concepts for describing trust-worthiness differs between the qualitative and thequantitative research traditions. Within the tradi-tion of qualitative content analysis, use of con-cepts related to the quantitative tradition, such asvalidity, reliability and generalisability, is stillcommon (for example, Downe-Wamboldt, 1992;Olson et al., 1998; Shields and King, 2001). Inqualitative research the concepts credibility, de-pendability and transferability have been used todescribe various aspects of trustworthiness (forexample, Guba, 1981; Lincoln and Guba, 1985;Patton, 1987; Polit and Hungler, 1999; Bergand Welander Hansson, 2000). However, Long andJohnson (2000, p. 31) propose that validity andreliability have ‘the same essential meaning’ irre-spective of research tradition and nothing is gainedby changing labels. In our paper, we suggest ap-plication of concepts linked to the qualitative tra-dition when reporting findings of studies usingqualitative content analysis. Even though we sep-arate the aspects of trustworthiness, they shouldbe viewed as intertwined and interrelated.

Credibility deals with the focus of the researchand refers to confidence in how well data andprocesses of analysis address the intended focus(Polit and Hungler, 1999). The first question con-cerning credibility arises when making a decisionabout the focus of the study, selection of context,participants and approach to gathering data.Choosing participants with various experiences in-creases the possibility of shedding light on the re-search question from a variety of aspects (Patton,


1987; Adler and Adler, 1988). In our illustrations,interviewees’ various genders and ages, and ob-servers with various perspectives, contributed to aricher variation of the phenomena under study.Selecting the most appropriate method for datacollection and the amount of data are also impor-tant in establishing credibility. The amount of datanecessary to answer a research question in acredible way varies depending on the complexity ofthe phenomena under study and the data quality.

Another critical issue for achieving credibility isto select the most suitable meaning unit. Meaningunits that are too broad, for example, severalparagraphs, will be difficult to manage since theyare likely to contain various meanings. Too narrowmeaning units, for example, a single word, mayresult in fragmentation. An exception to this iswhen one or several words represent a symbol ormetaphor. In both cases there is a risk of losingmeaning of the text during the condensation andabstraction process. Illustrating how meaningunits, condensations and abstractions are madefacilitates judging credibility of the findings (seeFigs. 1–3).

Credibility of research findings also deals withhow well categories and themes cover data, thatis, no relevant data have been inadvertently orsystematically excluded or irrelevant data in-cluded. Credibility is also a question of how tojudge the similarities within and differences be-tween categories. One way to approach this is toshow representative quotations from the tran-scribed text. Another way is to seek agreementamong co-researchers, experts and participants.

There are various opinions about the appropri-ateness of seeking agreement. Sandelowski (1993,1998) argues that, since multiple realities existthat are dependent on subjective interpretations,validation among co-researchers, experts and par-ticipants is questionable. Even though we agreethat reality is multiple and subjective, we defendthe value of dialogue among co-researchers. Theintent here is not merely to verify that data arelabelled and sorted in exactly the same way, but todetermine whether or not various researchers andexperts would agree with the way those data werelabelled and sorted (Woods and Catanzaro, 1988).Participants’ recognition of the findings can also bean aspect of credibility. It is not, however, aquestion of verification but rather a question ofconfirmability.

Another aspect of trustworthiness is depend-ability. According to Lincoln and Guba (1985, p.299), dependability ‘seeks means for taking intoaccount both factors of instability and factors ofphenomenal or design induced changes’, that is,

the degree to which data change over time andalterations made in the researcher’s decisionsduring the analysis process.

When data are extensive and the collection ex-tends over time, there is a risk of inconsistencyduring data collection. On one hand, it is importantto question the same areas for all the participants.On the other hand, interviewing and observing is anevolving process during which interviewers and ob-servers acquire new insights into the phenomenonof study that can subsequently influence follow-upquestions or narrow the focus for observation. Theextent to which judgements about similarities anddifferences of content are consistent over time can,as in our illustrations, be addressed by an open di-alogue within the research team.

Trustworthiness also includes the question oftransferability, which refers to ‘the extent towhich the findings can be transferred to othersettings or groups’ (Polit and Hungler, 1999, p.717). The authors can give suggestions abouttransferability, but it is the reader’s decisionwhether or not the findings are transferable toanother context.

To facilitate transferability, it is valuable to givea clear and distinct description of culture andcontext, selection and characteristics of partici-pants, data collection and process of analysis. Arich and vigorous presentation of the findings to-gether with appropriate quotations will also en-hance transferability.

There is no single correct meaning or universalapplication of research findings, but only the mostprobable meaning from a particular perspective. Inqualitative research, trustworthiness of interpre-tations deals with establishing arguments for themost probable interpretations. Trustworthinesswill increase if the findings are presented in a waythat allows the reader to look for alternative in-terpretations.

Reflections

When discussing meaning and use of concepts,procedures and interpretation related to qualita-tive content analysis, it is valuable to considerwhether qualitative content analysis is a separatemethod or tool used within different forms ofqualitative analysis. On one hand, a method that isso inexact that it fits into different research fields,methodological approaches and data can be seenas merely a tool. On the other hand, it can be as-sumed that qualitative content analysis has specificcharacteristics and underlying theoretical assump-tions which need to be further illuminated.


One characteristic of qualitative content analy-sis is that the method, to a great extent, focuses onthe subject and context, and emphasises differ-ences between and similarities within codes andcategories. Another characteristic is that themethod deals with manifest as well as latent con-tent in a text. The manifest content, that is, whatthe text says, is often presented in categories,while themes are seen as expressions of the latentcontent, that is, what the text is talking about.

One way to understand the theoretical assump-tions underlying qualitative content analysis is torelate the method to communication theory asdescribed by Watzlawick et al. (1967). They stateaxioms concerning human communication thatcould shed light on the issue of interpretation. Oneaxiom is that ‘one cannot not communicate’(Watzlawick et al., 1967, p. 51). Texts based oninterviews and observations are shaped within aninteraction between the researcher and the par-ticipants and can be seen as a communication act.In every text there are messages to be interpretedand described. As soon as the analysis procedurebegins, ongoing communication between the re-searcher and the text is present. Another axiom isthat ‘every communication has a content aspectand a relationship aspect such that the latterclassifies the former and is therefore a meta-communication’ (Watzlawick et al., 1967, p. 54). Inour illustrations, categories are seen as repre-senting the manifest content, that is, the contentaspect, and themes are representing the latentcontent, which can be seen as the relationshipaspect.

‘Human beings communicate both digitally andanalogically’ is another axiom of Watzlawick et al.(1967, p. 66). Verbal communication is mainlydigital and easily transcribed into a text while non-verbal communication is mainly analogical and of-ten put at a disadvantage in the transcriptionprocess. However, meaning is partly created byhow a message is communicated, that is, the voiceor implied feeling that emerges from the reading ofthe text (Downe-Wamboldt, 1992). Therefore,when transcribing interviews and observations intotext, it is valuable to notice silence, sighs, laugh-ter, posture, gestures etc., as these may influencethe underlying meaning. Watzlawick et al. (1967,p. 59) have also formulated the axiom that ‘thenature of a relationship is contingent upon thepunctuation of the communicational sequencesbetween the communicants’. Dividing the text intomeaning units is a way of punctuating the ongoingcommunication in a text and is important for bothmanifest and latent content when beginning andending a meaning unit.

Another aspect of interpretation is that a textalways involves multiple meanings and the re-searcher’s interpretation is influenced by his or herpersonal history. Since the researcher is often theone who collects the data as well as the one whoperforms the analysis, the question of the re-searcher’s qualifications, training and experiencesis important (Patton, 1990). In qualitative contentanalysis interpretation involves a balancing act. Onone hand, it is impossible and undesirable for theresearcher not to add a particular perspective tothe phenomena under study. On the other hand,the researcher must ‘let the text talk’ and notimpute meaning that is not there.

Learning and teaching how to analyse texts is adelicate matter in nursing education. Qualitativecontent analysis can be a valuable method forstudents when attending a research class for thefirst time due to the opportunity to perform theanalysis at various degrees of difficulty. Analysingcontent close to the text, that is, the manifestcontent, can be a suitable starting point. With in-creasing knowledge and ability students may ad-vance to interpret the underlying meaning, that is,the latent content, on various levels of abstraction.

In conclusion, our paper is intended to be used innursing research and education and to contributeto a debate on qualitative content analysis. In or-der to clarify the underlying assumptions of quali-tative content analysis, we suggest using conceptsrelated to qualitative research when describing theresearch procedure and measures to achievetrustworthiness. Moreover, we apply communica-tion theory as a way to address the issue of inter-pretation and clarify the underlying assumptions ofqualitative content analysis.

Acknowledgements

We are grateful to our colleagues at the Depart-ment of Nursing for fruitful reflections that helpedus to clarify our thoughts and for valuable sugges-tions for making the message clearer.

References

Adler, P.A., Adler, P., 1988. Observational techniques. In:Denzin, N.K., Lincoln, Y.S. (Eds.), Collecting and InterpretingQualitative Materials. Sage Publications Inc., Thousand Oaks,London, New Delhi, pp. 79–109.

Barrosso, J., 1997. Social support and long-term survivors ofAIDS. Western Journal of Nursing Research 19 (5), 554–582.

Baxter, L.A., 1991. Content analysis. In: Montgomery, B.M.,Duck, S. (Eds.), Studying Interpersonal Interaction. TheGuilford Press, New York, London, pp. 239–254.


Berelson, B., 1952. Content Analysis in Communication Re-search. The Free Press, Glencoe, Illinois.

Berg, A., Welander Hansson, U., 2000. Dementia care nurses’experiences of systematic clinical group supervision andsupervised planned nursing care. Journal of Nursing Manage-ment 8 (6), 357–368.

Burnard, P., 1991. A method of analysing interview transcripts inqualitative research. Nurse Education Today 11 (6),461–466.

Burnard, P., 1996. Teaching the analysis of textual data: anexperiential approach. Nurse Education Today 16 (4),278–281.

Cavanagh, S., 1997. Content analysis: concepts, methods andapplications. Nurse Researcher 4 (3), 5–16.

Coffey, A., Atkinson, P., 1996. Making Sense of Qualitative Data.Complementary Research Strategies. Sage Publications Inc.,Thousand Oaks, London, New Delhi.

Downe-Wamboldt, B., 1992. Content analysis: method, applica-tions, and issues. Health Care for Women International 13(3), 313–321.

Feeley, N., Gottlieb, L.N., 1998. Classification systems forhealth concerns, nursing strategies, and client outcomes:nursing practice with families who have a child with chronicillness. Canadian Journal of Nursing Research 30 (1), 45–59.

Findahl, O., H€oijer, B., 1981. Text- och inneh�allsanalys. En€oversikt av n�agra analystraditioner. (Swedish) (Text- andcontent analysis. A review of some analysis traditions). SRPublik- och Programforskning, Stockholm.

Graneheim, U.H., Norberg, A., Jansson, L., 2001. Interactionrelating to privacy, identity, autonomy and security. Anobservational study focusing on a women with dementia and‘behavioural disturbances’, and on her care providers.Journal of Advanced Nursing 36 (2), 256–265.

Guba, E.G., 1981. Annual review paper: criteria for assessing thetrustworthiness of naturalistic inquiries. Educational Com-munication and Technology: A Journal of Theory, Researchand Development 29 (2), 75–91.

Kondracki, N.L., Wellman, N.S., Amundson, D.R., 2002. Contentanalysis: review of methods and their applications in nutri-tion education. Journal of Nutrition Education and Behaviour34 (4), 224–230.

Kovach, C.R., 1991. Content analysis of reminiscences of elderlywomen. Research in Nursing & Health 14 (4), 287–295.

Krippendorff, K., 1980. Content Analysis. An Introduction to itsMethodology. The Sage Commtext Series, Sage PublicationsLtd., London.

Latter, S., Yerrell, P., Rycroft-Malone, J., Shaw, D., 2000.Nursing, medication education and the new policy agenda:the evidence base. International Journal of Nursing Studies37 (6), 469–479.

Lichstein, P.R., Young, G., 1996. My most meaningful patient.Reflective learning on a general medicine service. Journal ofGeneral Internal Medicine 11 (7), 406–409.

Lincoln, Y.S., Guba, E.G., 1985. Naturalistic Inquiry. SagePublications Inc., Newbury Park, London, New Delhi.

Lundman, B., Norberg, A., 1993. Coping strategies in peoplewith Insulin-Dependent Diabetes Mellitus. The DiabetesEducator 19 (3), 198–204.

Long, T., Johnson, M., 2000. Rigour, reliability and validityresearch. Clinical Effectiveness in Nursing 4 (1), 30–37.

Mertens, D.M., 1998. Research Methods in Education andPsychology. Integrating Diversity with Quantitative andQualitative Approaches. Sage Publications Inc., ThousandOaks, London, New Delhi.

Mishler, E., 1986. Research Interviewing. Context and Narrative.Harvard University Press, Cambridge, MA, London.

O’Brien, B., Relyea, J., Lidstone, T., 1997. Diary reports ofnausea and vomiting during pregnancy. Clinical NursingResearch 6 (3), 239–252.

Olson, M.S., Hinds, P.S., Eurell, K., Quargnenti, A., Milligan, M.,Foppiano, P., Powell, B., 1998. Peak and nadir experiencesand their consequences described by pediatric oncologynurses. Journal of Pediatric Oncology Nursing 15 (1), 13–24.

Patton, Q.M., 1987. How to use Qualitative Methods in Evalu-ation. Sage Publications Inc., Newsbury Park, London, NewDehli.

Patton, Q.M., 1990. Qualitative Evaluation and Research Meth-ods, second ed. Sage Publications Inc., Newsbury Park,London, New Dehli.

Polit, D.F., Hungler, B.P., 1991. Nursing Research. Principlesand Methods, fourth ed. J.B. Lippincott Company, Philadel-phia, New York, Hagestown.

Polit, D.F., Hungler, B.P., 1999. Nursing Research. Principlesand Methods, sixth ed. J.B. Lippincott Company, Philadel-phia, New York, Baltimore.

Sandelowski, M., 1993. Rigor or rigor mortis: the problem ofrigor in qualitative research revisited. Advances in NursingScience 16 (2), 1–8.

Sandelowski, M., 1998. Focus on qualitative methods. The call toexperts in qualitative research. Research in Nursing & Health21 (5), 467–471.

Shields, L., King, S.J., 2001. Qualitative analysis of the care ofchildren in hospital in four countries. Part 1. Journal ofPediatric Nursing 16 (2), 137–145.

S€oderberg, S., Lundman, B., 2001. Transitions experienced bywomen with fibromyalgia. Health Care for Women Interna-tional 22 (7), 617–631.

van Manen, M., 1990. Researching Lived Experience. HumanScience for an Action Sensitive Pedagogy. The University ofWestern Ontario, Ontario.

Watzlawick, P., Beavin Bavelas, J., Jackson, D.D., 1967.Pragmatics of Human Communication. A Study of Interac-tional Patterns, Pathologies and Paradoxes. W.W. Norton &Company, New York, London.

Weber, R.P., 1990. Basic Content Analysis, second ed. Series:Sage University Papers. Quantitative Applications in theSocial Sciences, vol. 49. Sage Publications Ltd., London.

Woods, N.F., Catanzaro, M., 1988. Nursing Research. Theory andPractice. The C.V. Mosby Company, St. Louis, WashingtonDC, Toronto.

qualitative content analysis

Documents