Why Bysios?

Operational area is the collection and evaluation of patient satisfaction questionings in craze hospitals, rehabilitation centers and/or therapy facilities for drug-addicts and dependence-gotten sick.
A goal of the program is to make the collection of the questionnaire data, based on the paper form, fast.
The evaluation is easy and fast to handle.
The output of the program is portable and might be easily integrated in other documents.
The program might be adapted to other questions.

Sprechen Sie deutsch? Do you speek german?

Install

Login as administrator and check, whether Java is already installed:
- Call Start execute cmd /x
- Type at the prompt
  java -version
  If you see
  
  "The command "java" is misspelled or can't be found."
  
  you havee to install Java.
- If you see
  java version "1.5.0_03"
  Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_03-b07)
  Java HotSpot(TM) Client VM (build 1.5.0_03-b07, mixed mode, sharing)
  
  then Java is already installed in the correct Version (1.5).
  "1.5" is crucial at the beginning
  "1.5.1" or "1.6" or higher is also ok.
  Continue at step 3.
- If you see
  java version "1.4..."
  ...
  or 1.3 or even smaller version numbers, you have a somewhat outdated version of Java. In most cases there is no problem in replacing an older version by a newer one. Ask your IT-administrator if unsure.
Java-Installation
- If you received an usb-disk (probably: USB-Disk F:\), you will find a directory "install", and therein a file "jre-1_5_0_03-windows-i586-p.exe". A doubleclick starts the installation. You are asked to accept the license, choose 'typical installation' and 'next'; at last 'Finish'. The Windows-PC has to be rebooted (not my fault).
bysios-installation:
- (again: be administrator) Copy the folder 'bysios' from the 'install'-dir to the target you like. This will normally be 'C:\Programs'.
- Release the folder (inkluding all subfolders) for the planned User:*)
  In the Windows-Explorer click withe the right button the tab 'security'.
  Choose "complete access" for all needed users / accept.
  Strictly speaking: You need write access to the file 'antworten.txt' and for the folder, where the reports go; inside the bysios-folder and in the demo-folder.
- Login as User.

*) If you are experienced in Windows-administration, and like to use an elaborated rights management, you may use all restrictions Windows offers. Just the data-collector needs strictly write-access to "antworten.txt", which might be placed whereever you like - you just have to modify the startscripts accordingly. If the user like to modify them too, they might create their own ones whereever they like. have a look at technicals

Collection

The data is collected in order of appearance on the quetionnaire.
A serial-number for the questionnaire is generated automatically thereby which is to be registered to control purposes on the questionnaire.
Multiple Choice questions limit the datenerfasser during the input to the possible values.
Per questionnaire side is indicated a side in the acquisition program, what permits easy control whether all questions became to transfer.
Screenshot acquisition program (page 1)

View Paperform (Part, page 1)

One can read off the coding with the multiple Choice questions
("1, 6, 3, 5...")
and thus transfer them fast into the program.
by the coding of the questions with an answer index
directly on the printed form
the transmission of the values is particularly simply and fast
details:
Start the acquisition script:
In the program directory, click erfassung.bat
The first question for the house is obligatorial. While no house is choosen, it's impossible to proceed to the next page (because some questions depend on the house).
The month is set by the program to the actual month. If you change the value, because you transfer the february-form in march, the following forms will use the same month.
The serial-ID is counted automatically and should be transfered to the paper form. This way you may check and correct later.
With the arrow right --> you proceed to the next page. With arrow left <-- you may step back to page 1, but not to former ... . On the last page, you step forward to the next ... by 'ok'. Later on the data is only to be modified by editor - not very userfriendly. Every ... is saved to disk by 'ok', and after the last ... when you hit 'esc' to leave the program.
Sometimes you will find pages which are monotonously answered with the same index - i.E. 5. Here the turbo-field comes handy, reducing your work. When you insert a number, let's say '5' into the field, and leave it (by mouse or keyboard (hitting the buttons is triggering a 'focus lost' too), all fields (where possible) are set to '5' on that page. Exceptions are Fields on page 1, and of course questions, where no answer 5 is available. The turbo-field is automatically cleared.
This technique is even helpful, if just nearly all answers are given monotonously. First set all answers to X, and switch few exceptions later. Try it yourself, to see whether it is helpful for you.
Did you fill in something by accident, and realize it early enough, so clear the field (delete, backspace, mouse) to avoid the filling with the value.
To test the programm, just copy the whole directory to a different location, and start it there. So you can get familiar with it and play around with the unused turbo-field.

Evaluation

For evaluation there is an extra modul:
The filters are easily adopted for different values, like different or a different count of groups of age, houses, and so on.
This might be reached by changing the questions-file.
More filters are only possible by changing the software.
By controlling by a batch-file, evaluations might be run in rows, which run automatically.

Evaluation examples

example (text)

Plus:
- small filesize
- for testing very useful
- might easily be further processed automatically
Minus:
- week graphics
- not easily processed by hand

View:

=============================================================================
6. Welches ist Ihr Hauptsuchtmittel?
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
graph		relativ	absolut	Antwort
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
..........	     8%	     11	(fehlt)
..........	     0%	      0	Medikamente
XXXX......	    44%	     52	illegale Drogen
XXXXX.....	    55%	     66	Alkohol
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
Durchschnitt: 2,6	gültige Antworten: 118
=============================================================================

7. Für meine Abhängigkeitserkrankung wurde die richtige Reha-Einrichtung ausgesucht?
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
graph		relativ	absolut	Antwort
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
..........	     0%	      0	(fehlt)
..........	     0%	      0	trifft überhaupt nicht zu
..........	     3%	      4	trifft eher nicht zu
X.........	    10%	     13	trifft teils zu, teils nicht
XX........	    28%	     37	trifft eher zu
XXXXX.....	    58%	     75	trifft voll und ganz zu
 -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -  -
Durchschnitt: 4,4	gültige Antworten: 129
=============================================================================

Example (HTML)
- Plus:
  - small filesize
  - high portability
  - for ad-hoc Evaluations most useful
- Minus:
  - Most Printers don't support page-breaks in HTML
- View
Example (pdf)
- Plus:
  - Device independent (Acrobat-Reader, printer)
  - Index allows fast navigation
  - well suited for printing
- Minus:
  - big file sizes
  - hardly further processable
- View
Example (pdf.chart)
- Plus:
  - nicest graphics
  - Index allows fast navigation
- Minus:
  - big file size
  - long texts are cut
  - only useful with colour printers
- View

Remarks to the interpretation

Percentvalues are calculated in two steps. All filtered questionaires are counted. The percentvalue for missing answers is based on the number of all forms. If there is an explicit option 'nicht teilgenommen' (not available), this is alos calculated against all Forms. All other percentvalues are generated without missing answers (and without, while present: 'nicht teilgenommen'). Bildschirmfoto Interpretation

The average is generated while leaving out missing Answers, (and maybe without 'nicht teilgenommen'), and is in itself a misleaded teechinque for qualitative answers, because you shouldn't calculate averages on ordinal values. The first column is coded as 0, the second as 1 and so on. An average of 4.6 says, that most answers are 4 or 5 - something between 'good' and 'very good'.
Maybe the values are useful to compare the values between different months, but you shouldn't threat them as meaningful values on their own.
Just more boo-boo is the average on sex and main drug, but for technical reasons there is no distinction in handling of the values.

A graphical Overview about programm and files is to be seen here: program and files

File formats

This informationen might be useful for you for different reasons. Maybe you like to proceed your data with different tools like Spreadsheets, SPSS or R. Or you did an error on data input, and already saved the row. Then you may open the file antworten.txt in the editor, and change it there. Or maybe you like to modify the formulations of text - answers and questions. You might add new own questions or remove existing ones. Then you need to take care on consistency of data in between, and know the format of course. Modification and insertion of questions and answers isn't easy, but possible. The first 6 Questions can't be modified, because the filters in the program are hard coded. Maybe a later version will be more flexible in that direction.

fragen.txt

This file is used by collection
Every question is a row.
For every question:

Columns are separated by tabs.
The 1st column is an enumeration of questions.
The 2nd column marks, for which house the question is quested.
- A 0 means: all houses.
- A single number (i.e. 3) means, for the house (department) with index 3 from question 1 the question is made. (Haus Hendrix)
- More numbers mean, for all these Houses the question is given (1,2,4: all houses except Haus Hendrix).
The 3rd column indicates on which page the question is printed on the paper form, and in the programm.
Column 4 shows the typ of question.
- r creates a collection of radiobuttons in the collection program.
- d creates a date-field which is only a month and a year field for free input
- z creates a counter, which generates serial IDs.
- i creates an integer-field, as you can see by question: 'Zum wievielten Mal füllen Sie den Bogen?' for free input.
Column 5 contains the questiontext. It might be shorter than in the evaluation-part, because it is only needed to identify the question - not needed for printing.
Column 6 and following are the answer texts . That's the main difference to fragen.txt

Most questions are of type r and are used for all houses.
Example:

0	1	r	In welcher Einrichtung der Demoklinik befinden Sie sich?	Haus Zappa	Haus Joplin	Haus Hendrix	Haus Morrisson
1	1	d	Berichtsmonat
2	1	z	Fragebogen-ID
3	1	i	Zum wievielten Mal in dieser Einrichtung füllen Sie den Bogen aus?
4	1	r	Sind Sie weiblich oder männlich?	1	2
5	1	r	Zu welcher Altersgruppe gehören Sie?	1	2	3	4	5	6
6	1	r	Welches ist Ihr Hauptsuchtmittel?	1	2	3
7	1	r	Für meine Erkrankung wurde die richtige Einrichtung ausgesucht?	1	2	3	4	5
8	1	r	Meine Meinung wurde bei der Auswahl der Einr. zu wenig berücksichtigt.	1	2	3	4	5
9	2	r	Ich habe zuwenig ärztliche Betreuung erhalten	1	2	3	4	5

atexte.txt

This file is needed by the evaluation.
For every question there is a row.

For every question:

Columns are separated by tabs.
The first question is an enumeration for all questions
The second column specifies, for which houses the question is asked.
- A 0 means all houses.
- A single number (i.e. 3) means, the question is asked for the house with number 3 from question 1. (Haus Hendrix)
- More numbers mean, for all this houses the question is asked (1,2,4: all houses except Haus Hendrix).
The 3rd column names the page where the question is told.
The 4th column indicates the questiontyp and controls therefor intern handling of the answers by the program.
The 5th column contains the question text, in the way, how it is printed by the report.
The 6th and following columns are indizees of answers.

Most questions are of typ r and asked for all houses.
Example:

0	1	r	In welcher Einrichtung der Demoklinik befinden Sie sich?	Haus Zappa	Haus Joplin	Haus Hendrix	Haus Morrisson
1	1	d	Berichtsmonat
2	1	z	Fragebogen-ID
3	1	i	Zum wievielten Mal in dieser Einrichtung füllen Sie den Bogen aus?
4	1	r	Sind Sie weiblich oder männlich?	weiblich	männlich
5	1	r	Zu welcher Altersgruppe gehören Sie?	bis 20	21-30	31-40	41-50	51-60	61 und darüber
6	1	r	Welches ist Ihr Hauptsuchtmittel?	Medikamente	illegale Drogen	Alkohol
7	1	r	Für meine Abhängigkeitserkrankung wurde die richtige Reha-Einrichtung ausgesucht?	trifft überhaupt nicht zu	trifft eher nicht zu	trifft teils zu, teils nicht	trifft eher zu	trifft voll und ganz zu
8	1	r	Meine Meinung als Patientin/ Patient wurde bei der Auswahl der Einrichtung zu wenig berücksichtigt.	trifft überhaupt nicht zu	trifft eher nicht zu	trifft teils zu, teils nicht	trifft eher zu	trifft voll und ganz zu
9	2	r	Ich habe zuwenig ärztliche Betreuung erhalten.	trifft überhaupt nicht zu	trifft eher nicht zu	trifft teils zu, teils nicht	trifft eher zu	trifft voll und ganz zu

antworten.txt

This file is created by collection and gets appendet, and is read by the Evaluation.
For every form there is a row.

For all answers:

Columns are separated by spaces.
The 1st column is the house, corresponding with row 1 in fragen.txt and atexte.txt.
The 2nd column is month and year, where the form was filled.
The 3rd column is the serial number, usefull for control and correction
The 4. column contains the answer for the question 'Zum wievielten Mal füllen Sie den Bogen aus?'
The 5. column und all followings, contain the answer for the question for that column, which is the row in the file atexte.txt
With other words: Column i corresponds with row i from atexte.txt and fragen.txt.
An answer j corresponds with column 5+j from atexte.txt and fragen.txt.
For example '3' in column 8 means, for question 8 the 3rd answerpossibility was choosen, which is "trifft teils zu, teils nicht zu" for "Meinung wurde zuwenig berücksichtigt;

Example (stripped to 20 columns):

1	9.2005	1	1	2	3	2	4	1	0	0	5	4	4	4	4	4	4	3	1	1	1	1	1	1
1	9.2005	2	1	1	1	2	3	3	3	2	1	1	1	3	2	2	2	2	1	1	1	2	1	1
1	9.2005	3	2	1	1	2	5	5	4	3	2	5	3	5	5	5	5	6	6	6	5	6	1	1
1	9.2005	4	1	1	2	2	4	1	2	5	5	5	5	3	5	4	6	6	1	1	1	4	1	1
1	9.2005	5	1	2	1	2	4	2	4	2	0	4	3	2	2	2	6	6	1	6	5	6	4	1
1	9.2005	6	1	1	2	2	3	2	4	2	2	2	3	2	4	3	4	4	5	1	1	4	1	1
1	9.2005	7	1	2	2	2	5	0	5	1	1	1	1	1	1	3	4	6	1	1	1	5	1	1
1	9.2005	8	1	1	2	2	5	1	1	3	4	5	5	4	4	4	5	6	1	1	1	4	1	1
1	9.2005	9	2	1	2	2	5	1	2	3	4	5	5	3	5	4	6	6	1	6	1	6	5	1
1	9.2005	10	2	2	2	2	4	4	3	4	3	4	3	3	4	4	5	6	1	1	1	5	1	1

technicals

You may take influence on the programs Erfassung and Auswertung by starting it with params.
Not everybody is interested in such features, so you don't need to use them.
But maybe you may save yourself some work this way, or make working with restricted user rights more easy.

Erfassung:
Normally you start Erfassung with the batch-file
erfassung.bat

Which calls

java -jar erfassung.jar

, which has the same consequence like starting by mouse click or by

java -jar erfassung.jar -f fragen.txt -a antworten.txt
, because the programm looks for the data (fragen.txt, antworten.txt) in the current directory.
To save the answers in a different file (i.E. the user has write-access there), the path to the answers can be given that way:
java -jar erfassung.jar -f fragen.txt -a "C:\Dokumente und Einstellungen\frank\bysios\antworten.txt"
If you don't like to remember such a Command, you will add it into a new batch-file.� Since you need write access for such a batch file too, you would store it somewhere below "My Files". Then you have to give the whole path to the jar-file and to fragen.txt, the question file. I.E.:
java -jar C:\Programme\bysios\erfassung.jar -f C:\Programme\bysios\fragen.txt -a "C:\Dokumente und Einstellungen\frank\bysios\antworten.txt"
or, more accurate:
BYSIOS_HOME=C:\Programme\bysios\
BYSIOS_DATA="C:\Dokumente und Einstellungen\frank\bysios\"
java -jar %BYSIOS_HOME%\erfassung.jar -f %BYSIOS_HOME%\fragen.txt -a %BYSIOS_DATA%antworten.txt
Auswertung
For Auswertung it is similar, which you start by
auswertung.bat
.
This calls

java -jar auswertung.jar -f atexte.txt -a antworten.txt -o auswertung.pdf

Without further options, where atexte.txt (answer texts) and antworten.txt (choosen answers) are, and which output file to produce, you may fill the textfields to specify the paths, or open a file chooser dialog to specify them by hitting the buttons.
At least for the first two files it's convenient to have the fields prefilled. This is why the batch file selects them in the current working directory. An output filename is preselected too. You may overwrite that file allways, or choose a different name everytime, or move/rename the output after each session.
Auswertung has much more options, which is useful, if you like to produce a lot of evaluations partly automatical. If you run the program with the single option -? or with invalid options, the following help messages are printed to stdout:
```
Usage:
	java -jar auswertung.jar [Optionen]
		Optionen:
			-? Diese Hilfe anzeigen
			-f Fragentextedatei (z.B.: atexte.txt)
			-a Antwortendatei (z.B.: antworten.txt)
			-o Output (z.B.: auswertung.html)
			-d Datumsfilter (MM.JJJJ-MM.JJJJ)
			-g Geschlechtsfilter ([m|w])
			-h Hausfilter (z.B.: 1, 12, 135, ...)
			-j Altersgruppe ([J]ugendlichkeit)  (z.B.: 1;2;3 =untere ~, 4;5;6 obere Altersgruppen)
			-s Suchtmittelfilter ([p|i|a] (Polytoxikoman, illegale Drogen, Alkohol)
			-v Starte Viewer zur Anzeige der Ergebnisse (xpdf/ acroreader, firefox/ opera)
			-w Wievieltes Mal der Ausf�llung ([1, 2, 3, ...]) (keine Gruppenbildung)
			-x eXecute (true) automatisch mit diesen Werten ausf�hren.
				sinnvoll, um das Programm per Script zu steuern.
				
```
- (An english Version will only be available on demand)
- For the -f and -a option see above.
- The -o option controls not just the name of the outputfile Ausgabedatei, but its format too.
  - A .txt creates a simple textfile, which is useful for testing purposes - forget about it.
  - A .pdf creates a pdf-file. This is the recommendet option, and most useful for printing or sending by mail.
  - A .html creates a html-file. This is much smaller than a PDF, but it's easy to forget about the 0.dot.png - graphicfile when sending it by mail to someone else. If you don't see a graphic, it happened to you. Since pagebreaks in HTML-files are mostly set by the printer randomly, pages often get cut at bad positions. But for viewing in the browser and publishing in the intranet, the html-option is very useful.
  - If you choose the double ending .chart.pdf the graphics will be made by the jFreeChart lib. Those graphics will differ in width and layout and the labeling is sometimes cut unfortunately. In case of heavy demand this format might get improved.
- The following options are useful for partly automated mass evaluations, or if you frequently use the same filter:
  - -d MM.JJJJ-MM.JJJJ chooses a timespan (-d date) i.e. 04.2005-03.2006, or 01.2006-01.2006. Avoid blanks between the dates.
  - -g [m|w] (Geschlecht=Sex, m=maskulin, w=weiblich=feminin
  - -h Choose a house or a section -h 2. For multiple houses -h 23
  - -j N[;N...] -j Jahre=years for the age-group. -j 2 for the second age group. -j 1;2;3 for the 3 youngest age groups
  - -s [m|i|a] Suchtmittel=prefered Drug: p=polytoxikoman, i=illegal drugs, a=alkohol. Two prefered drugs aren't groupable.
  - -w [1| 2|3|...] wievieltes Mal=how often did the client participate in the questionaire. No grouping here too.
- The following options control the grade of automation:
  - With -x true (eXecute) the program is not just started with these values filled in, but starts immediately and produces the result file. This is convenient for creating multiple evaluations in row. With appropriate knowledge in shell scripting this can be used to produce job queues, for instance to send the results immediately per mail.
  - -v VIEWER can be used in connection with the -x option to open a viewer after producing the results, to view the results. However, you have to give the whole path to the viewer and mask blanks in the pathname.
    I.e.:
    - java -jar auswertung.jar -f atexte.txt -a antworten.txt -o auswertung.pdf -x true -v "C:\\Programs\\Adobe\\Acrobat 5.0\\Reader\\AcroRd32.exe"
    - java -jar auswertung.jar -f atexte.txt -a antworten.txt -o auswertung.html -x true -v "C:\\Programs\\Firefox\\firefox.exe"
    Such commands are of course preferably placed into a script.

Performance

Testet with Hardware:
- 2000 Mhz CPU (Centrino)
- 512 MB Ram memory
- (a middle of the road PC of 2005)
Testet while filtering:
- 5 of 6 houses
- sex: male
- 5 of 6 age groups
- a time filter without effekt

Performance in seconds:

questions	questionaieres	output format
questions	questionaieres	txt	html	pdf	chart
60	100	1	1	6	9
	1.000	2	2	8	12
	10.000	4	3	10	12
	100.000	30	28	38	39
	1.000.000

Interpretation:
For a small amount of data (< 1000 questionaires) the number of questionaires is not important, since most time is spent on layout.
About 50% of the questionaires survived the filtering.
With 100.000 questionaires the program performes well. That would be 10 evaluations per year for 100 years with 100 clients per evaluation.
Testing 1 million rows reached the RAM-limit of the machine.
This number shows that you might get a more poweful machine before reaching that limit.
Testing without filter proceedet in comparable times. The costs of filtering is compensated by the smaller number of questionaires to evaluate.
For systems with slower CPU the results should be transferable, as long as the machine doesn't start swapping because of limited RAM. This might only be expected on very old machines which try to handle far more than 10 000 data rows.

Price

Bysios is put under the license GPLv3
You don't need to pay anything to use bysios .
You may run it on as much desktops, laptops, PDAs or mobiltelephones as you like.
You may even modify the source code or find somebody to modify it.
And you may redistribute that modified program (just don't call it bysios anymore, and don't expect me to support your modified version).
The words of the whole license are to be found here: GPLv3 .
If you like me to make the program fit to your needs, and give support, you have to pay something.
These costs are subject to the individal case, and depend on the expected amount of work and support.
To give you an idea - calculate 600 € for the initial typical adjustment, installation support and a first year of support. Calculate 100 €. for every extra year of support.
As mentioned before - prices might be fixed in a different way. Depending on the number of houses/sections of your facility, it might be worse.
If you read the documentation carefully, and have good inhouse know how in computing and java, you might get it done on your own without any support and therefore without costs.
The template questionaire as word file (on demand: Open Office, perhaps something else) belongs to the program.
On license or costs-questions write an email.
To protect me from spam you have to modify this address . Replace the -at- with @. Thank you.

Download

download:

name	suggestion	size	dl with 57k modem	dl with 1MB DSL
bysios.tar.bz2	Linux	5.5 MB	≈ 15 min	≈ 45s
bysios.zip	Windows	5.7 MB	≈ 15 min	≈ 45s

Thanks & third party licenses

This program uses some free libraries in adherence to their licenses. The license for the program itself Lizenz für das Programm selbst finden Sie hier: GPLv3

itext a library to create PDF files. MPL license
pngencoder a library to create png graphics. LGPL license
jFreeChart a library to create chart graphs. license
Additional thanks to Frau A. Heußer and Herrn H. Brömer, which contributet to the project by their time, mostly to semantic questions.

Top | enail | authors homepage | counter/

Why Bysios?

Sprechen Sie deutsch? Do you speek german?

Install

Collection

Evaluation

Evaluation examples

example (text)

Plus:

Minus:

View:

Example (HTML)

Plus:

Minus:

View

Example (pdf)

Plus:

Minus:

View

Example (pdf.chart)

Plus:

Minus:

View

Remarks to the interpretation

File formats

fragen.txt

atexte.txt

antworten.txt

technicals

Erfassung:

Auswertung

Performance

Price

Download

Thanks & third party licenses

Menu

german version