Release notes Dub InterViewer version 6.3 - February 2006

 
Coding
Set filter on date range.
Warning log appears when chosen item does not exist
Text-questions can now be coded.
Distribution improvement: In a Multiple Response Question there was the total number in the bottom. This total is split into ’Total number or respondents’ (that got the question) and ‘Total number of answers’.
Distribution improvement. In a Multiple Response question you find next to the overview of given responses also how many respondents did not get the question. This is removed. The distribution function is meant to see which % of the respondents fall in the other category.
Answers can now be deselected when coding a Multiple Response question (internally a field ‘Coded’ is set to True).
Translate verb items in re-coder. Text can be posted to for example Babble fish and the result to the request can be put into the screen as a default.
When using, while recoding, the option 'Find Respondent', when the code of the respondent doesn't exist, a pop up message appears, 'Respondent xyz not found'.
 
Analysis
Filter on quota.
Grouping of bockrotated items.
In ‘Response Statistics’ the screened out respondents can be seen under ‘Screened out respondents, by question’.
Open items of Single Response and Multiple Response questions are now also shown in ' All verbatims'.
 
Question types
Properties in items of text-questions (i.e. E-mail, password etc).
Check domain of an E-mail address.
Record order in which multiple response answers are given.
Retrieve answer items for Single- and Multiple responses from an enumeration in Dub Knowledge.
 
Project Management
You can select a project/survey type (detailed information here).
Store start and end time for each project (detailed information here).
Store Nett N for each project (detailed information here).
New mainscreen that shows more information per project (ProjectLeader usertype)
  • Fieldwork start/end
  • Date/time last complete interview
  • Nett N (target)
  • Completed absolute and %
  • Nr of invites
  • Nr of respondents that opened mail (not a strict value, depends on respondents settings)
  • Nr of respondents that started interview
  • Nr screened out
  • Buy
  • Incidence rate
  • Response % (Started/Invited)
  • Avg time to complete
    • Overview of nr of mails sent out :
  • Per week
  • Per month
  • Per day
  • For a day, per project/schedule
    • Message is shown when a project with sample is added to ManageFrames without access control.
    Old mainscreen shows ‘NO DATASET’ instead of ‘xxx’, to make the users better aware that they forgot to recreate the dataset.
    In ‘Correct interview’ the data does not have to be filed in before listing all interviews that can be corrected.
    Fixed: Removing a respondent form ManageFrames, whose partial data has been saved to buffer and dataset, was not completely removed from buffer and dataset.
     
    System Management sysman login of Manage Frames
    You can see system usage Total/ Per project/ Per month / Per week.
    Overviews of pending schedules Today/ Tomorrow and later.
    Overview of nr of mails sent out see 'Project Management'.
    Access rights:
    • Assign projects to projectgroups
    • Assign users to usergroups
    • Which usergroups are allowed to see which projectgroups
    • What (basic) actions can be taken
    (detailed information here).
    Restart system counter every day.
    Save daily performance info in table DubInt-> System Performance
  • Request made
  • Total requsttime
  • Avg time for a request (KPI)
  • Successful writes to Dub Knowledge
  • Unsuccessful writes to Dub Knowledge
  • Successful entity-id generates
  • Unsuccessful entity-id generates
  • Timeouts
    • Restart system counter every day.
     
    Expressions in interviews
    Possibility to set language in an expression.
    Check domain of an E-mail address.
    New Sample-variable knowledge Entity is available as a system-variable in the questionnaire.
     
    CATI
    New login screen for interviewers: login by code/username + password (detailed information here)
    Project selection in separate screen - for interviewers (detailed information here.)
    New possibility for interviewers to select language and quota cells (like location), (detailed information here.)
    Login in briefing mode (like ‘take a break’ screen) (detailed information here.)
    No auto logoff from ‘ take a break’ and ‘ briefing mode’.
    Breakdown of interview duration into Contact, Screening, Talk and Review. InterviewDuration contains total..
    Now possibility to show ‘ No answer given’ in CATI (was already available in Data Entry).
    Subcases can be made to return to interviewer who made them.
    New way how to manage workstations, the old terminal code is replaced by workstations (detailed information here.)
     
    WAPI
    Use templates, stylesheets and images in library. No need anymore to copy them into projectfolder.
    Dynamic selection of template name, by setting system-variable “Template” in expression.
    Dynamic selection of template name, by using wildcards in the name:
  • <#KnowledgeEntity>, <#PanelID> Entity-id in Dub Knowledge (store in sample in DubQuest)
    • >
  • <#LangID> Universal Dub-code of language (0=Dutch, 1=English, etc)
  • <#Lang2LC> Generic 2-character identifier for language (ne, en, de etc)
  • <#Lang4LC> Generic 4 – character identifier for language
  • Example : Template:=’Template_<%~Lang2LC%>_<%~PanelID%>’
    • Option in open questions and text questions to prevent automatically setting the focus.
    URL in end form now redirects into top frame instead of in frame where questions are shown.
     
    Improvements in project templates.
    Are centralized, so you don't retain a version per project (for templates and login pages)
    Naming of templates and login pages with language and panel-id. Each respondent automatically gets the right login and template. (Template_NL_1.htm for Dutch respondent in panel 1), using an expression (Template := 'Template_<%~Lang%>_<%~PanelID%>');
     
    Data Entry
    Re enter the respondent number at the end of interview, for verification purposes. This can simply be done using a numeric form, expression and skip. An expression can override the initial value (Respondent := CorrectResp).
    Use other options for start screen, 1 and 9 can't be used as respondent-numbers.
    Show respondent number in top of screen.
    When going back, 'None of these' (in text and numeric) is not selected when it should.
     
    Connection to Dub Knowledge
    Faster connection to SQL -server, old way used too much overhead.
    Fixed: Generating new Dub Knowledge Entity-id was done in testinterviews.
    Besides 'read from Dub Knowledge' and 'write to Dub Knowledge' also a function 'clear feature in Dub Knowledge' was added.
    Fixed: Save interviews in ManageFrames: Questionnaire-feature in DK was not set to '6'(Completed partially).
    The titles of the results in the Dub Knowledge search form were only in English. Improvement they can be shown in the language of the questionnaire.
    Questionnaire-feature-id´s are stored internally to reduce number of reads from DK-database.
    Store internal UTF-8 format in UNICODE format into Dub Knowledge (2.3.0 and higher).
    Update ' The longest questionnaire filled (seconds)' system feature value.
     
    Improvements in Quota
    New structure for storing and calculating quotacells (faster).
    Optimistic quotacells in CATI.
    Non closing quotacells .
    When stopping an interview for a continuation appointment (CATI), the quota cell(s) are stored and before resuming you can decide if it is still needed.