The Code-A-Text Integrated System for the Analysis of Interviews and Dialogues (C-I-SAID) consists of an integrated suite of software programmes developed to facilitate the study of documents, usually based upon interviews or dialogues but can include documents such as questionnaires.  C-I-SAID has its most obvious applications in psychology, psychotherapy, sociology, medicine and market research but also has uses in wider areas such as law and the humanities - particularly in theatre

C-I-SAID offers a set of unique features which allow the depth of analysis, normally associated with a programme dedicated to qualitative analysis, within an analytic framework which is quantitative in orientation. Thus open ended coding, using comments and textual annotations, is accompanied by rating scales that can be categorical or numerical in format. There are also sections within the programme such as the lexicon (largely devoted to thematic content analysis) and the acoustic manager (devoted to the measurement of the volume, pitch and speed of speech) that generate statistics. The main outputs from C-I-SAID are reports, tables and charts which can be accompanied by a range of statistics these can be used to describe the data in tandem with qualitative methods.

C-I-SAID has been designed to enable the investigator to engage closely with their subject matter by making the original data, which may be a sound or video recording or a set of texts available at any stage in the analysis. Thus a paragraph in a report, a cell in a table or a point in a chart will usually be linked to their source data. A few mouse clicks can take you from a statistical description to the video, sound and text on which that was based.

The source data  is usually some form of human communication. This normally occurs in a situation where the participants can see and hear each other and use language to communicate. C-I-SAID enables studies using these different communication channels by allowing the investigator to use video, audio and text. Because it is sometimes impractical to use all three C-I-SAID makes it as easy to study a single channel often sound without video or text or, more commonly, text alone as in the  analysis of questionnaires.

The changes are found in three main areas.

These new features make C-I-SAID a unique programme that provides the depth of analysis associated with qualitative studies within a quantitative analytic environment. It is a flexible tool, which can as easily be used for the analysis of questionnaires (from data input to final report) as it can for the detailed analysis of a recorded or video taped dialogue. These different approaches are illustrated in the demonstration section of the User Guide

C-I-SAID bases its analyses on the documents within a project. A document can consist of sound, text and video. Each document is subdivided into speech units which in turn consist of segments. The segment is the normally the unit which is analysed. The user defines a segment which may be a speaking turn, a section of text consisting of a set number of words, or even a period of sound broken by a silence. In many cases C-I-SAID will automatically create the segment structure.  With Version 3 it is now possible to use the speech unit or the document as the analytic unit.

C-I-SAID has facilities for recording or copying audio files on to the computer and a number of features for transcription, including the capacity to slow sound down, insert silences into text, or to stop playback for a specific period of time. C-I-SAID contains a spell checker with British and American English dictionaries, of more than 200,000 words, and the facility to add dictionaries from other European languages .

C-I-SAID uses hypertext links to form associations between different parts of a document. Such links mean that the transcription and the media files are synchronised and the user can move between them with ease. These links, which are made by the user, also provide the facility to move between codes and the underlying source data.

Although C-I-SAID is specifically designed to analyse dialogues and has special facilities to accommodate different speakers with functions that enable the investigator to study the process sequentially it is, however, by no means restricted to this format. It is possible to code a media file directly without any transcription of sound, indeed the programme could easily be used for the study of music. 

C-I-SAID is not restricted to the study of conversation. C-I-SAID uses the convention that a normal document without speakers contains a single speaker and can automatically format such a document accordingly. Indeed from entering a Text Document to obtaining a full content analysis can take a matter of seconds.

C-I-SAID can also import data formatted as tab delimited files. The Data Editor introduced in version 3 provides a powerful and flexible means of editing and creating these files. Its text windows mean that questionnaire and other forms of data can be typed straight into C-I-SAID. The import functions can automatically create new documents and add the required scales and codes from a tab delimited file.  

C-I-SAID creates an index of every word in a project. Words can be brought together in to word groups which relate to concepts that interest the investigator. C-I-SAID contains two internal dictionaries each of which contains more than 50 word groups involving more than 5,000 words which can be used as the basis of user defined groups. Word groups can also be created by importing them from external sources or generating them from the documents in a project.  

A comment is a piece of text added by the user to a segment. Comments are stored apart from any texts in the segments to which they apply. C-I-SAID creates a separate index of words in the comments which can be searched in the same way as the text to which it is attached.

An annotation is a word entered into the main body of the text which functions as a code. Annotations might be used to indicate features such as tone of voice. C-I-SAID attaches an index of annotations to the comments index.

User Created

Rating Scales are created by the user to try and capture the underlying essence within the source data. C-I-SAID allows the user to create numerical, ordinal or nominal scales. Rating scales can be either categorical or numerical. Categorical scales contain codes in discrete categories. E.g. A scale of Anxiety might have three ordered categories High, Medium, Low a scale Food might have categories such as Meat, Fish, Cheese. By contrast numerical scales only contain numbers.

Scales and codes are added to a project through the Rating Scale Editor. In the editor the user can add instructions as to how a scale should be coded. These are displayed when the user elects to code a segment on a scale.

There are five methods for manually coding rating scales.

Auto Coding

Computer Generated

Acoustic Scales

C-I-SAID can search a project using any of the coding methods described above. Apart from the normal Boolean operators C-I-SAID has a number of unique operators which allow for complex searches. There is no limit to the number of phrases that can be built into a search. Output from a search is the segments meeting the search criteria which can be presented with their properties (text, media, codes) as reports, tables or spreadsheet. The output from a search can also be directed to a Rating Scale or be used to create a new document.  

Searches can specify which documents and speakers to include. Searches of words and word groups can be filtered so that only segments with specific codes are included. The user can move the focus of a search from the segment to the speech unit. Word group searches can be focused within syntactical units. For instance when searching for the co-occurrence of two concepts the user can determine that the search criteria will only be met if the occur within the same phrase or sentence rather than the same paragraph (segment).

Searches that output to reports take the form of documents which include the text of the segments that meet the criteria and links to the media file. The lexical structure of the output such as word counts and frequencies with comparisons to those segments which don't meet the search criteria are also available 

Can be applied to all words and word groups in a project. There are currently 17 different functions in this part of C-I-SAID. Many of these have multiple representations.

These include

Many functions also have graphical outputs.

Rating Scale Analysis provides functions for analysing and graphically representing the Rating Scales in a project. Each output from Rating Scale Analysis is accompanied by a chart, clicking on which will take the user back to the source data.

The main features of Rating Scale Analysis are

A variety of methods exist for manipulating document structure. These include creating documents based on

Rating scales can be manipulated by computing a new scale

C-I-SAID supports the following