Questions about summarydata and data

Back

Questions about summarydata and data

Author	Message
ai293	ai293 posted 2 Months Ago Topic Details Share Topic
Group: Forum Members Posts: 7, Visits: 68	Hi! I'm running my first study on Inquisit (please bear with my beginner skills), but I'm running into issues with data storage. Specifically, I have a few questions/concerns: 1) my demographics survey, when it downloads, has data in more columns than what is labeled (e.g., there are 8 columns and there's data up until column. I'm not sure what's causing this issue, or what this data is referring to. I'm also not sure why the survey is downloading in a separate file from the rest of the "summarydata". 2) I'm not sure what "win" is and why the cells in the summary data file sometimes hold the value "win" (e.g., block num, dropdown race). 3) I'm not sure what columns to include in my data file, and most of them remain empty when I run monkey. From the examples I've seen, most people have "stimulus number" or "expression ", but I don't understand how people are making these decisions and how they design their overall data file. 4) How do you decide whether to use this format: surveypage.fc_text8.response or textbox.fc_text8.response (for reference, fc_ text is both a textbox and a surveypage, with the textbox inside the survey page)? Is there a difference in output? 5) Do I need to add recorddata for each piece of data? I've seen examples using ontrialend (this is from the willingness to forgive likert scale in your library): /ontrialend = [list.ratings.appendl em(likert. scenario.response list.latencies.appenditem(likert. scenario.latency)] and I've also seen (from your library - anagrams task) /ontrialend = values.anagramCondition = "easy" How does the "ontrial end" shape the data collection, and what are codes like "appenditem" doing? Do I need to use it for my likert scale too (I don't have a likert scale in this script, but I have it in another script I'm developing)? 6) For checkboxes, do I need to use checkboxes.race.checked OR checkboxes .race.response? I will include my code for the data and summary data below, and I will attach my code (as well as the data files that I'm referring to). Also, for clarity, my aim is to structure the data files in a way that collects all the open-ended responses (+demographics) for all questions. I'm aware that many of my questions reflect a lack of knowledge related to Inquisit, and I've spent a lot of time reading past questions on this forum, reading the handbook, playing around with examples, and hours of trial and error, but I still can't seem to figure these out, so any help would be appreciated. Thanks! <summarydata> / columns = (script. filename, script.startdate, script. starttime, script.subjectid, script.groupid, script.elapsedtime, computer. platform, dropdown.english.response, dropdown.latino.response, dropdown.relationships.response, dropdown.religion.response,dropdown.religiosity.response, dropdown.sex.response, dropdown.grad_year.response, checkboxes.race.checked, radiobuttons.consentdecision.response, radiobuttons.followupconsent.response, surveypage.dttext1.response,surveypage.dt_text2.response, surveypage.dt_text3.response, surveypage.fc_text1.response, surveypage.ic text2 response,surveypage.ic esponse, surveypage.fc text4.response, surveypage.fc text5.response surveypage.fc text6. response, surveypage.fc_text7.response, surveypage.c_lexi&.re e surveypage.fc texi9 response, surveypage.sn_text1.response, surveypage.sn text2.res surveypage. sn text3.response, s surveypage.sn text5.response, surveypage.sn_text6.response, survevpage.sn text7.response, surveypage.sn_text8.response,surveypage.sn_ text.response, surveypage.sn text10.response, surveypage.sn_text11.response, textbox.sn_q11_text.response, textbox.major.response, textbox.age.response, textbox.dt_q1_text.response, textbox.dt_q2_text.response, textbox.englishage.response, textbox.dt_q3_text.response) / labels = true / separatefiles = false </summarydata> ‹data> / separatefiles = false / labels = true / columns = (script. filename, script.startdate, script.starttime, script subjectid, script. groupid, script.elapsedtime, computer.platform, surveypage.dt_text2.response, surveypage.dt_text3.response) </data> Attachments friend_labeling_s0_trial3_summary.iqdat (32 views, 5.00 KB) demographics_survey.iqdat (30 views, 2.00 KB) friend_labeling_s0_trial3_raw.iqdat (32 views, 41.00 KB) friend_labeling_study0_trial3.iqx (27 views, 37.00 KB)
	Reply Like 2
Dave	Dave posted 2 Months Ago ANSWER Post Details Share Post
Group: Administrators Posts: 13K, Visits: 104K	+x ai293 - 10/30/2024 Hi! I'm running my first study on Inquisit (please bear with my beginner skills), but I'm running into issues with data storage. Specifically, I have a few questions/concerns: 1) my demographics survey, when it downloads, has data in more columns than what is labeled (e.g., there are 8 columns and there's data up until column. I'm not sure what's causing this issue, or what this data is referring to. I'm also not sure why the survey is downloading in a separate file from the rest of the "summarydata". 2) I'm not sure what "win" is and why the cells in the summary data file sometimes hold the value "win" (e.g., block num, dropdown race). 3) I'm not sure what columns to include in my data file, and most of them remain empty when I run monkey. From the examples I've seen, most people have "stimulus number" or "expression ", but I don't understand how people are making these decisions and how they design their overall data file. 4) How do you decide whether to use this format: surveypage.fc_text8.response or textbox.fc_text8.response (for reference, fc_ text is both a textbox and a surveypage, with the textbox inside the survey page)? Is there a difference in output? 5) Do I need to add recorddata for each piece of data? I've seen examples using ontrialend (this is from the willingness to forgive likert scale in your library): /ontrialend = [list.ratings.appendl em(likert. scenario.response list.latencies.appenditem(likert. scenario.latency)] and I've also seen (from your library - anagrams task) /ontrialend = values.anagramCondition = "easy" How does the "ontrial end" shape the data collection, and what are codes like "appenditem" doing? Do I need to use it for my likert scale too (I don't have a likert scale in this script, but I have it in another script I'm developing)? 6) For checkboxes, do I need to use checkboxes.race.checked OR checkboxes .race.response? I will include my code for the data and summary data below, and I will attach my code (as well as the data files that I'm referring to). Also, for clarity, my aim is to structure the data files in a way that collects all the open-ended responses (+demographics) for all questions. I'm aware that many of my questions reflect a lack of knowledge related to Inquisit, and I've spent a lot of time reading past questions on this forum, reading the handbook, playing around with examples, and hours of trial and error, but I still can't seem to figure these out, so any help would be appreciated. Thanks! <summarydata> / columns = (script. filename, script.startdate, script. starttime, script.subjectid, script.groupid, script.elapsedtime, computer. platform, dropdown.english.response, dropdown.latino.response, dropdown.relationships.response, dropdown.religion.response,dropdown.religiosity.response, dropdown.sex.response, dropdown.grad_year.response, checkboxes.race.checked, radiobuttons.consentdecision.response, radiobuttons.followupconsent.response, surveypage.dttext1.response,surveypage.dt_text2.response, surveypage.dt_text3.response, surveypage.fc_text1.response, surveypage.ic text2 response,surveypage.ic esponse, surveypage.fc text4.response, surveypage.fc text5.response surveypage.fc text6. response, surveypage.fc_text7.response, surveypage.c_lexi&.re e surveypage.fc texi9 response, surveypage.sn_text1.response, surveypage.sn text2.res surveypage. sn text3.response, s surveypage.sn text5.response, surveypage.sn_text6.response, survevpage.sn text7.response, surveypage.sn_text8.response,surveypage.sn_ text.response, surveypage.sn text10.response, surveypage.sn_text11.response, textbox.sn_q11_text.response, textbox.major.response, textbox.age.response, textbox.dt_q1_text.response, textbox.dt_q2_text.response, textbox.englishage.response, textbox.dt_q3_text.response) / labels = true / separatefiles = false </summarydata> ‹data> / separatefiles = false / labels = true / columns = (script. filename, script.startdate, script.starttime, script subjectid, script. groupid, script.elapsedtime, computer.platform, surveypage.dt_text2.response, surveypage.dt_text3.response) </data> 1) <survey> elements always produce a separate data file, because data output for surveys is different than data output for blocks and trials. Survey data is in "wide" format. Block / trial raw data is in "long" format. 2) If columns are mismatched, this typically means you first collected some data with a script that recorded X set variables in the data file, then you later changed that script and added or removed some variables and/or changed their order. So, you have two (or more) differently formatted sets of data. Start with a fresh data file, better yet, record every run in separate data files per /separatefiles = true in your <survey>, <data> and <summarydata> elements. Merge data files as needed later. On the web, everything is in separate files anyway. Selected files are merged when you download them. 3) You should record all data you may possibly need for analysis. Better to record more than you end up needing than too little. It's not possible to get data you have never recorded. You must know what you want to do, therefore only you can decide what you need and what you don't need. 4) A <surveypage> has no response property. The question elements displayed by the surveypage have response properties. If you want to log the response property of a textbox element, you reference the textbox's response property: textbox.textboxname.response. 5) /ontrialend executes logic at the end of a trial. This has nothing to do with "shaping" data recording or /recorddata. As for what things like appenditem() do, there is a large documentation where you can look things like this up: https://www.millisecond.com/support/docs/v6/html/language/functions/appenditem.htm There is also a Programmer's Manual, which you should read and work through in full: https://www.millisecond.com/support/Inquisit%20Programmer's%20Manual.pdf 6) Either or both, depending on what you want.
	Reply Like 3