Search the Community
Showing results for tags 'matrix coding'.
Found 3 results
Hello, I am trying to do something that I think should be simple, but I am getting the wrong results and I can't work out why. I have attached a word document with screen shots to show what I am doing, in the hope that someone can tell me what I am doing wrong - LGough matrix coding problem.docx I have an NVivo project with one source - a dataset of survey respondents. Each row is a unique respondent - there are 477 rows, therefore 477 unique respondents. There are many columns containing the answers to a range of closed and open ended questions. I have created a "respondent" parent node which contains a child node for each respondent . Each child node has attribute values according to the respondent's answers to the closed-ended questions. Two of the closed ended questions asked are "Q1 What cohort are you in" (undergraduate, postgraduate etc) and "Q2 what is your method of study" (full-time, part-time etc). All 477 respondents answered these two questions. I want to essentially create an Excel-style Pivot Table which counts how many respondents fall into each combination (pair) of attributes - e.g., how many full-time undergraduates are there, how many part-time postgraduates are there etc. I know what the answer should be as I can create the pivot table in Excel, but I want to be sure I can get the same answer in NVivo as I am very new to using it and want to make sure I understand what I am doing. Whatever I try in NVivo (including every option under "cell content" in the matrix coding query) gives me the incorrect numbers. Can anyone tell me what I am doing wrong, it is driving me insane! Thanks.
Hi all, I'm fairly new to NVivo and am using it to help me organise articles in a systematic review. I decided to code each source on a number of characteristics (rather than creating classifications- potentially wrong decision but I can change that next time). It would be great if I could now run matrix coding queries to identify the sources that have been coded at multiple nodes. That said, when I coded each source, in many cases the content of each article was automatically coded to the node as well, so for each source coded at a node there are up to 20 references. Therefore, when I run a coding query I get a whole heap of text. I have run reports to identify sources coded at each individual node. Is there a way I can do something similar to get a list of sources coded at multiple nodes? I hope this makes sense. I feel like i'm speaking another language! Many thanks in advance, Kari
Hi everyone, I've got a question about NVivo9. I haven't found any answer in the forum, but could have missed it. In NVivo9. I imported my dataset from Excel. It contains answers to closed and open-questions. I coded 3 open-ended questions in a hierarchic node structure. Let's say: - question_1 ----- A1 ----- B1 ----- C1 ----- D1 - question_2 ----- A2 ----- B2 ----- C2 ----- D2 - question_3 ----- A3 ----- B3 ----- C3 ----- D3 Some answers were coded under 2 lower level categories. For example: one participant's answered to question 1 was coded under A2 and B2 (not sure whether this makes any differnce). In the classification, I have Respondents, where each closed-questions for the respondents are classified. This classification is done using the following node: - Respondents ---- Repondent1 ---- Respondent2 ---- Respondent3 ---- and so on... Each of these nodes has all open-ended question for the respective respondent. Each of them is classified as the "Respondent" classification (this was done automatically using NVivo9) I also have 1 extra node: - Data ---- question_1 ---- question_2 ---- question_3 each of these nodes has all answers for each open-ended question I can create Matrix coding query using the Respondent classification vs question_1, question_2 or question_3 subnodes. What I haven't figured out how to do is to generate Matrix coding query to compare question_1 with question_2 for instance. I've tried several different options and I'm really struggling now. What I want to analyse is, for example, is how many people were coded as A1 and also A2 and so on. If possible, I want to generate 1 query to cross tab A1, B1, C1 and D1 with A2, B2, C2 and D2. Thanks for your help, it's much appreciated!