| AmCAT Version |
|---|
| This page describes a feature in AmCAT 3.3 |
| View other version: 3.3 - 3.4 - 3.5 |
The output option Article List in the Query search is the second of three main functions of automatic content analysis in AmCAT. The Article List function displays your search results as a table in which each row is one article in AmCAT, and each column is a variable that says something about that article, such as headline, length or for each search string the amount of hits in that article. (This is NOT a cross tab, which you create using the graph/table output option). You can export this table from AmCAT for further use in Excel, SPSS or another statistics program, where you can merge it with a file containing manual codings (use article IDs to link the cases). Another useful feature of the Article List is its option to display Keyword in Context for each search keyword, which is one of the variables you can choose to put in the columns of the Article List table. This can help you improve your search strings, as we will explain below.
First, select your articleset, time interval and media, and enter one or more search strings in the Keyword Search field. If you want to generate a table containing all articles without searching for a specific concept, leave the Keyword Search field empty. Select the output option 'Articlelist' and click on the 'Query' button. By default, the articlelist table has the variables Article ID, Date, Headline, Medium ID and Medium in the columns. Click on the ‘5 selected’ field to the right of ‘Columns:’ to add or remove variables. ‘Keyword in context’ is the second variable on this list.
The 'Keyword in Context' variable shows your search term in the specific context of the article, that is, the words to the left and to the right of the search term in the text. As such, it can help you determine whether your search string generates good results, or if it needs to be refined. An example might make this more clear. Let’s say we want to find articles about the British Prime Minister David Cameron. His last name seems not extremely common, so perhaps just the word ‘cameron’ is sufficient as a search string. A search using the Keyword in Context variable in an Article List can help us see if this is indeed the case.
We want to look at a selection of articles from the newspaper The Guardian from 2015, in project 888, for which we want to use the full time interval. The search term ‘cameron’ is entered in the Keyword Search window, and we make an articlelist with only article ID and Keyword in Context as the variables.
After clicking ‘Query’ AmCAT generates a table in which the top rows look like this
The search term ‘cameron’ is listed in the middle of the table, and to its left and right the rest of the sentence in which the word ‘Cameron’ occurs. As you can see, not all Camerons in this list refer to Prime Minister David Cameron. There is also a Geoff Cameron in the list, who is an American soccer player and not relevant to what we are searching for. The search term ‘cameron’ thus has to be refined further, either by adding conditions to it (such as ‘“prime minister” AND cameron’) or by excluding unwanted findings (‘cameron NOT “geoff cameron”’). In this way, Keyword in Context helps you make clear where your search string needs more refinement.
Click on the Export button on the top right of the Article List table to export this table as a csv, Excel or SPSS file.
AmCAT
