Formats

Formats for sending data by eMail

Send your coded SUMI data in an eMail message to jzk@uxp.ie by the agreed date. Your data will be analysed and returned by eMail within one working day. Please don't send me unsolicited data! This is always ignored and deleted.

The basic data format is a CSV (Comma Separated Value) subset, which can be transmitted as:

plain text in an eMail message or
a MIME-encoded attatchment as a .CSV or a .TXT file.

The CSV format is said to contain records, one record per line. Each record contains fields separated by commas. The rules for the SUMI subset of CSV are as follows:

Records are separated by a newline character.
The basic separator required between fields is a comma (please note, not a semicolon!)
If any fields in your data contain commas, the field should be enclosed in double quotation marks.
The format can also handle double quotation marks in a field, but because implementations differ on how this is coded, please check that none of your fields have included double quotation marks (replace with single quotation mark if you find any doubles.)
There must always be the same number of fields in each record. An empty field is signalled by having no characters between the two comma separators.

Most spreadsheets will help you prepare a CSV format file. There are some examples of correctly coded files lower down on this page.

The SUMI questionnaire data are coded as follows:

'Agree'.......	1
'Undecided'	2
'Disagree'....	3

If a user omits a question then that question is coded as a zero (0 -- not a capital O).

Each user's SUMI responses are coded as the first field in each record, with a space every ten responses. After the line of 50 digits, you put a comma, and any additional data fields that you wish to keep with the data.

The first line of the file should give some information regarding what the data is, and when it was gathered.

Here is an example of a very simple correctly prepared file (example1):

Simple SUMI data created 20/04/2007 3113131221 3112131311 3312313111 1111332311 2121123133, aengus 2211112231 1221221131 2121131321 3112111121 1213111211, brid 3113111333 1112131311 3111111331 3111111131 1113112311, carmel 3111211223 1111231221 3313222213 1122312221 1121213233, dounat 3113111223 1221131321 2312221211 2111231221 1111212123, ethne

Here is an example of a correctly prepared file with many fields after the SUMI data (example2):

SUMI data with many fields per record created 20/04/2007 3213213231 3213213231 3212321232 3231232132 3123123123,Following the vehicles in our fleet,1,2,it's very complete,the menu is very tiny writing 3123123132 3212321232 1323132123 1232123212 1232321232,"To keep track of our lorries, trucks, and vans",1,1,The quality of the information afforded,"Really, no complaints"

The above example has:

50 coded SUMI questions
one freeform comment
two numeric (forced-choice) comments
two free-form comments

It could have come from the standard online SUMI form at sumi.uxp.ie/en . Please note that the number of fields in each record must be the same for the entire dataset. Null fields are communicated by no character between the comma separators.

You may wish to add a line which explains what each field in the dataset means (always useful for record keeping.) You do this by making the first field a null field. Here is the above example, with labels in the fields (example3):

SUMI data with many fields per record created 20/04/2007 ,What for,How Often,How long,Best feature,Improvement 3213213231 3213213231 3212321232 3231232132 3123123123,Following the vehicles in our fleet,1,2,it's very complete,the menu is very tiny writing 3123123132 3212321232 1323132123 1232123212 1232321232,"To keep track of our lorries, trucks, and vans",1,1,The quality of the information afforded,"Really, no complaints"

You'll see how the fields line up if you cut and paste this into a spreadsheet.

You can download CSV files and TXT versions of these three examples from the list below (use your browser's back arrow to get back here.)

[SUMI Homepage] [Top of Page]