| Part ofa series on |
| Research |
|---|
 |
Research strategy |
| Philosophy portal |
Content analysis is the study ofdocuments and communication artifacts, which are defined as texts. Examples of texts include photographs, speeches, and essays. Social scientists employ content analysis as a method of examining patterns in communication in a replicable and systematic manner.[1] One of the key advantages of using content analysis to analyse social phenomena is their non-invasive nature, in contrast to simulating social experiences or collecting survey answers.
Practices and philosophies of content analysis vary between academic disciplines. They all involve systematic reading or observation oftexts or artifacts which areassigned labels (sometimes called codes) to indicate the presence of interesting,meaningful pieces of content.[2][3] By systematically labeling the content of a set oftexts, researchers can analyse patterns of contentquantitatively usingstatistical methods, or usequalitative methods to analyse meanings of content withintexts.
Computers are increasingly used in content analysis to automate the labeling (or coding) of documents. Simple computational techniques can provide descriptive data such as word frequencies and document lengths.Machine learning classifiers can greatly increase the number of texts that can be labeled, but the scientific utility of doing so is a matter of debate. Further, numerous computer-aided text analysis (CATA) computer programs are available that analyze text for predetermined linguistic, semantic, and psychological characteristics.[4]
Content analysis is best understood as a broad family of techniques. Effective researchers choose techniques that best help them answer their substantive questions. That said, according toKlaus Krippendorff, six questions must be addressed in every content analysis:[5]
The simplest and most objective form of content analysis considers unambiguous characteristics of the text such asword frequencies, the page area taken by a newspaper column, or the duration of aradio ortelevision program. Analysis of simple word frequencies is limited because the meaning of a word depends on surrounding text.Key Word In Context (KWIC) routines address this by placing words in their textual context. This helps resolve ambiguities such as those introduced bysynonyms andhomonyms.
A further step in analysis is the distinction between dictionary-based (quantitative) approaches and qualitative approaches. Dictionary-based approaches set up a list of categories derived from the frequency list of words and control the distribution of words and their respective categories over the texts. While methods in quantitative content analysis in this way transform observations of found categories into quantitative statistical data, the qualitative content analysis focuses more on the intentionality and its implications. There are strong parallels between qualitative content analysis andthematic analysis.[6]
Quantitative content analysis highlights frequency counts and statistical analysis of these coded frequencies.[7] Additionally, quantitative content analysis begins with a framed hypothesis with coding decided on before the analysis begins. These coding categories are strictly relevant to the researcher's hypothesis. Quantitative analysis also takes a deductive approach.[8] Examples of content-analytical variables and constructs can be found, for example, in the open-access databaseDOCA. This database compiles, systematizes, and evaluates relevant content-analytical variables of communication and political science research areas and topics.
Siegfried Kracauer provides a critique of quantitative analysis, asserting that it oversimplifies complex communications in order to be more reliable. On the other hand, qualitative analysis deals with the intricacies of latent interpretations, whereas quantitative has a focus on manifest meanings. He also acknowledges an "overlap" of qualitative and quantitative content analysis.[7] Patterns are looked at more closely in qualitative analysis, and based on the latent meanings that the researcher may find, the course of the research could be changed. It is inductive and begins with open research questions, as opposed to a hypothesis.[8]
The data collection instrument used in content analysis is the codebook or coding scheme. In qualitative content analysis the codebook is constructed and improvedduring coding, while in quantitative content analysis the codebook needs to be developed and pretested for reliability and validitybefore coding.[4] The codebook includes detailed instructions for human coders plus clear definitions of the respective concepts or variables to be coded plus the assigned values.
With the rise of common computing facilities like PCs, computer-based methods of analysis are growing in popularity.[9][10][11] Answers to open ended questions, newspaper articles, political party manifestos, medical records or systematic observations in experiments can all be subject to systematic analysis of textual data.
By having contents of communication available in form of machine readable texts, the input is analyzed for frequencies and coded into categories for building up inferences.
Computer-assisted analysis can help with large, electronic data sets by cutting out time and eliminating the need for multiple human coders to establish inter-coder reliability. However, human coders can still be employed for content analysis, as they are often more able to pick out nuanced and latent meanings in text. A study found that human coders were able to evaluate a broader range and make inferences based on latent meanings.[12]
Robert Weber notes: "To make valid inferences from the text, it is important that the classification procedure be reliable in the sense of being consistent: Different people should code the same text in the same way".[13] The validity, inter-coder reliability and intra-coder reliability are subject to intense methodological research efforts over long years.[5]Neuendorf suggests that when human coders are used in content analysis at least two independent coders should be used.Reliability of human coding is often measured using a statistical measure ofinter-coder reliability or "the amount of agreement or correspondence among two or more coders".[4] Lacy and Riffe identify the measurement of inter-coder reliability as a strength of quantitative content analysis, arguing that, if content analysts do not measure inter-coder reliability, their data are no more reliable than the subjective impressions of a single reader.[14]
According to today's reporting standards, quantitative content analyses should be published with complete codebooks and for all variables or measures in the codebook the appropriate inter-coder orinter-rater reliability coefficients should be reported based on empirical pre-tests.[4][15][16] Furthermore, thevalidity of all variables or measures in the codebook must be ensured. This can be achieved through the use of established measures that have proven their validity in earlier studies. Also, thecontent validity of the measures can be checked by experts from the field who scrutinize and then approve or correct coding instructions, definitions and examples in the codebook.
There are five types of texts in content analysis:
Content analysis is research using the categorization and classification of speech, written text, interviews, images, or other forms of communication. In its beginnings, using the first newspapers at the end of the 19th century, analysis was done manually by measuring the number of columns given a subject. The approach can also be traced back to a university student studying patterns in Shakespeare's literature in 1893.[17]
Over the years, content analysis has been applied to a variety of scopes.Hermeneutics andphilology have long used content analysis to interpret sacred and profane texts and, in many cases, to attribute texts'authorship andauthenticity.[3][5]
In recent times, particularly with the advent ofmass communication, content analysis has known an increasing use to deeply analyze and understand media content and media logic. The political scientistHarold Lasswell formulated the core questions of content analysis in its early-mid 20th-century mainstream version: "Who says what, to whom, why, to what extent and with what effect?".[18] The strong emphasis for a quantitative approach started up by Lasswell was finally carried out by another "father" of content analysis,Bernard Berelson, who proposed a definition of content analysis which, from this point of view, is emblematic: "a research technique for the objective, systematic and quantitative description of the manifest content of communication".[19]
Quantitative content analysis has enjoyed a renewed popularity in recent years thanks to technological advances, being fruitfully applied in mass and personal communication research. Content analysis of textualbig data produced bynew media, particularlysocial media andmobile devices has become popular. These approaches take a simplified view of language that ignores the complexity ofsemiosis, the process by which meaning is formed out of language. Quantitative content analysts have been criticized for limiting the scope of content analysis to simple counting, and for applying the measurement methodologies of the natural sciences without reflecting critically on their appropriateness to social science.[20] Conversely, qualitative content analysts have been criticized for being insufficiently systematic and too impressionistic.[20] Krippendorff argues that quantitative and qualitative approaches to content analysis tend to overlap, and that there can be no generalisable conclusion as to which approach is superior.[20]
Content analysis can also be described as studyingtraces, which are documents from past times, and artifacts, which are non-linguistic documents. Texts are understood to be produced by communication processes in a broad sense of that phrase—often gaining mean throughabduction.[3][21]
Manifest content is readily understandable at its face value. Its meaning is direct. Latent content is not as overt, and requires interpretation to uncover the meaning or implication.[22]
Holsti groups fifteen uses of content analysis into three basiccategories:[23]
He also places these uses into the context of the basic communicationparadigm.
The following table shows fifteen uses of content analysis in terms of their general purpose, element of the communication paradigm to which they apply, and the general question they are intended to answer.
| Purpose | Element | Question | Use |
|---|---|---|---|
| Make inferences about the antecedents of communications | Source | Who? |
|
| Encoding process | Why? |
| |
| Describe & make inferences about the characteristics of communications | Channel | How? |
|
| Message | What? |
| |
| Recipient | To whom? |
| |
| Make inferences about the consequences of communications | Decoding process | With what effect? |
|
| Note. Purpose, communication element, & question from Holsti.[23] Uses primarily fromBerelson[24] as adapted by Holsti.[23] | |||
As a counterpoint, there are limits to the scope of use for the procedures that characterize content analysis. In particular, if access to the goal of analysis can be obtained by direct means without material interference, then direct measurement techniques yield better data.[25] Thus, while content analysis attempts to quantifiably describecommunications whose features are primarily categorical——limited usually to a nominal or ordinal scale——via selected conceptual units (theunitization) which are assigned values (thecategorization) forenumeration while monitoringintercoder reliability, if instead the target quantity manifestly is already directly measurable——typically on an interval or ratio scale——especially a continuous physical quantity, then such targets usually are not listed among those needing the "subjective" selections and formulations of content analysis.[26][27][28][29][30][31][15][32] For example (from mixed research and clinical application), as medical imagescommunicate diagnostic features to physicians,neuroimaging'sstroke (infarct) volume scale called ASPECTS isunitized as 10 qualitatively delineated (unequal) brain regions in themiddle cerebral artery territory, which itcategorizes as being at least partly versus not at all infarcted in order toenumerate the latter, with published series often assessingintercoder reliability byCohen's kappa. The foregoingitalicized operations impose the uncreditedform of content analysis onto an estimation of infarct extent, which instead is easily enough and more accurately measured as a volume directly on the images.[33][34] ("Accuracy ... is the highest form of reliability."[35]) The concomitant clinical assessment, however, by theNational Institutes of Health Stroke Scale (NIHSS) or themodified Rankin Scale (mRS), retains the necessary form of content analysis. Recognizing potential limits of content analysis across the contents of language and images alike,Klaus Krippendorff affirms that "comprehen[sion] ... may ... not conform at all to the process of classification and/or counting by which most content analyses proceed,"[36] suggesting that content analysis might materially distort a message.
The process of the initial coding scheme or approach to coding is contingent on the particular content analysis approach selected. Through a directed content analysis, the scholars draft a preliminary coding scheme from pre-existing theory or assumptions. While with the conventional content analysis approach, the initial coding scheme developed from the data.
With either approach above, researchers may immerse themselves into the data to obtain an overall picture. A consistent and clear unit of coding is vital, with the choices ranging from a single word to several paragraphs and from texts to iconic symbols. Lastly, researchers construct the relationships between codes by sorting out them within specific categories or themes.[37]
| International | |
|---|---|
| National | |
| Other | |
