Jump to content

Pablo Gustavo Rodriguez

Trainers
  • Content count

    45
  • Joined

  • Last visited

Community Reputation

0 Neutral

About Pablo Gustavo Rodriguez

  • Rank
    Member
  • Birthday 06/21/1961

Contact Methods

  • Website URL
    http://www.analisiscualitativo.com.ar
  • ICQ
    0
  • Yahoo
    autenticopgr@yahoo.com.ar
  • Skype
    pablo.g.rodriguez

Profile Information

  • Gender
    Male
  • Location
    La Plata, Buenos Aires, Argentina
  • Interests
    Social anthropology, social economy, local development, poverty and cyberanthropology.
  1. Pablo Gustavo Rodriguez

    How does Nvivo handles cluster analysis?

    Marcus: Very much thanks for your answer. It meets my needs. I´ll read the paper you attached, and hope this helps too.
  2. I am writing my doctoral thesis and I´m worried about some methodological issues dealing with the way Nvivo handles cluster analysis. Exactly I wonder how it handles data. I don´t feel prepared to answer questions about this issue to my examination comitee if they would want to know. I don´t mean the basis of cluster analysis that can be found in a statistics manual, but the choices the software make after the user click the "finish" button. I would like to know the technical details about how Nvivo handles cluster analysis in order to be able to interpret results correctly as well as properly address ussues of validity. I´m talking about something like a white paper. Not the kind of explantion that can be found in the software help files or tutorial, but the kind of information about operations the programer used to write the code and algorytms that process the data. I mean what "choices" does the software do after I chose to work with resources/nodes and analyse words/coding similarity?. I would like to read an official QSR document about this issue that I can cite and refer to in case I would be asked, rather than a short informal post. For example, when the user chooses "word similarity" what words are considered? Every word o are some words ignored, like "blind words"?. And when the user chooses "coding similarity" what nodes are considered? All, including cases nodes, matrix nodes, relationships, query results and everything?. If the user wants to restrict the comparison to just some nodes, can it be done?. I wanted to do it so I created some temporal case nodes containing only the passages coded at the nodes I wanted to restrict the comparison to. Then I used these nodes to do the cluster analysis. What nodes are compared in the analysis in this case?. Also I would like to read what other users think of and/or address issues of vality of cluster analysis of qualitative data with Nvivo.
  3. After reading the discusion on the importance on memos in Nvivo Linkedin group I realized that there´s no way to know what text passages and/or how many are linked to a certain memo, from that memo. In the list view you can see what nodes are linked to a memo via "link to memo" and even open the memo, and also you can see what memos are linked to a node and also open the node. When reading a document or node content I can see what memos are linked to passages by "see also links" ticking the option to see those links in the "view" menu in the ribbon. But the reverse is not possible: when reading a memo there´s no way to see what passages. nodes or sources are linked to it from via "see also links". It would be useful to be able to see this information as well. Regards,
  4. Pablo Gustavo Rodriguez

    cluster analysis

    Hi Nina: I can´t give an answer because I can´t figure out yor project from your description. First of all I don´t understand how could you import excel spreadsheets as source documents to code, as Nvivo only import spreadsheets to process surveys. I don´t know what does PKI means and why do you have 250 nodes if you say you coded the 30 documents in 25 nodes. With this restrictions in my knowledge about your project I will try to help the best I can. Assuming you have text documents in your internals folder and that coding is already done, a cluster analysis on all 30 sources clustered by coding similarity is run including all nodes. You can´t choose to run the cluster analysis using only sub nodes or any particular subset of nodes. Nvivo uses all the nodes coding the source documents. If every document represent a unit of analysis (a health centre) the chart shows visually how near the centres are in terms of the themes that appear in their respective Progress Reports (the KPIs which I don´t know what represent). So it is an overall similarity. It's up to you to interpret what the chart means, based in the themes you used to code them. I don´t know of any "method" to do it. Just think of the questions you are trying to answer with this analysis and try to figure out how this chart answer some of them. Your intention to export the cluster analysis results to Excel to further analyze the "counts", seems to me to be grounded in cuantitative reasoning, and I can´t figure out what would be the use of it. Regarding the question "what is a good way to decide how many clusters to choose?", I would say the number of clusters that make sense. Comparing with cuantitative methods it is like choosing a significance level. It's your choice. Though you have to justify it. Sorry if it was of little help. Good luck!
  5. Pablo Gustavo Rodriguez

    Nvivo strange behaviur when coding pdf files

    Hi everyone: Though there are other topics about Nvivo problems in handling pdf files I preferred to open a new one because this implies a specific problem. The issue is that Nvivo doesn´t allways code the passage that I select, but sometimes ignore some of the highlighted words and also code words outside the highlighted passage. All my attempts to correct this mistakes were unsuccessful. I tried to code only the ignored words, and also to code a larger passage and later to uncode the surplus words. But it seems like Nvivo reads the words in a different sequence that the one one reads. I leave you a screenshot to illustrate this issue: I desired to code the first two words ("manera efectiva") and I selected them as part of the larger passage that ends with "international", but Nvivo wouldn´t code them. Besides, the second passage starting at paragraph 7, should end at the period but the last tree words don´t get coded along with the rest of the passage. Instead the lonely word "de", (wich I did not select) that got coded along with the preeceding passage. That´s really weird. This is just one example, but it happens in many passages. And the document has only one column. I hope this issue could be fixed in a next upgrade.
  6. Mucha suerte Cesar. Saludos desde Argentina.
  7. Pablo Gustavo Rodriguez

    NVivo 9.1

    I also applaud and welcome this improvement. I´m impressed and grateful. Now I can analyze Human Development Reports.
  8. Pablo Gustavo Rodriguez

    Printing a simple list of nodes

    Hi jdevlin: It worked for me just as Pat indicated. Make sure you took the second step: 2) Click on every + sign to expand any branch if you have sub-nodes, or else they won´t be printed. Good luck Pablo
  9. Hi Kath: Very much thanks for sharing this information with all of us. I´m sorry I didn´t answer before, but I didn´t because I don´t use social networks so they didn´t change my life at all. I appreciate privacy a lot so I don´t use them for my personal activities, and I either need them for research, so far. I would though if it were useful for my research purposes. For millons of people in the world life still occurs mainly offline and I work precisely with them. Nevertheless as I teach research methods at university I will recommend your texts to students interested in researching online. Cheers, Pablo
  10. Pablo Gustavo Rodriguez

    Coding trees for use with Nvivo 9

    Mine are in spanish
  11. Pablo Gustavo Rodriguez

    Coding trees for use with Nvivo 9

    Kath's steps are really good advices and reminded me of an article I read several years ago when I was doing my firts steps in qualitative analysis. Un fortunately this article, writen by Tom and Lyn Richards were not found online anymore. So I just digitalized my hard copy to make it accessible anew (I hope Lyn won't get angry with me for that). The article is: Richards, T., & Richards, L. ynn. (1995). Using Hierarchical Categories in Qualitative Data Analysis (training report). QSR International, Melbourne, Australia. Just click on the link to access the text and enjoy it. Pablo
  12. Pablo Gustavo Rodriguez

    Coding trees for use with Nvivo 9

    Hi alexquse: I completely agree with you, as I understand you have said you used a different tree structure for each one of your different studies. I also agree that we can discuss about the logic of coding and building tree structures. This is something different than using the same code tree for many researches. Every research is unique because it has unique objectives and questions. And the conceptual scheme one builds up responds to this specific questions. I think your examples clarify what I said. Thanks for that. Pablo
  13. Pablo Gustavo Rodriguez

    Coding trees for use with Nvivo 9

    Hello kdeanna: Your request is very difficult ti satisfy because it is founded in two false suppositions: 1) Certain code trees can be useful for many different researches no matter the subject under study 2) Some code trees work better with particular software of analysis. A coding tree is a conceptual framework and, as such, it comes in part from your theoretical framework and in part stems from your own work on your data. In consequence every single research is unique and develops a unique coding tree. Besides as far as I know there´s no restriction arising from software to develop a certain structure of codes you need. In a nutshell it is not a software matter implied in this issue but only a theoretical or methodological issue. So I think a good way to find some suggestions on possible categories of analysis useful for your own research is reading publications and thesis on the same or similar matter of study. You will find your own way to articulate some of this categories in a new and useful way with other categories developed by yourself and/or coming out from your data, depending on your methodological preferences. So Why don´t you tell us what are you working on? Pablo
  14. Pablo Gustavo Rodriguez

    Nvivo 9 - some words of warning

    Hi Kevin: There could be many reasons for your problem (an error While installing the software, a conflict with another program running, a disk error, a memory error, etc.). If there is something I appreciatte with Nvivo in spite of its frequent errors is an excellent user support service. Communicate with them and explain your problem. First, look at "My documents" folder for the error log files. Almost allways Nvivo crashes it creates an error log file in this folder. Its name is something like err2011-03-23T20-17-12.log where the firts numbers are the date the error occurred (in this example, 2011-03-23). This file contains clues for the user suport service to guess what the problem is. Attach this files to your message to support service. Tell them also : 1) what kind of system you have (Operating system and version, 32 or 64 bits, processor model, RAM available, etc.) 2) What other software were you running when the crash occurred 3) What was the task you were trying to perform when the crash occurred With this information they can help you. Some problems may be general, but other problems depend on the system characteristics. Good luck Pablo
  15. Pablo Gustavo Rodriguez

    using nodes to write up report

    Hello Bernard: There is not a way to automatically group your free nodes "conceptually". Only you know how to group them. I think it is correct to create parent nodes to hang this nodes from. You cannot create "children" nodes before parent nodes, as a children node is one that hangs from a parent. But you can create parent nodes and later paste corresponding free nodes as children.
×