Algemeen/Onderzoeksmethoden 2/het werk/2011-12/Group 1/results/comparison

Table

Concept	RWS	HK	JB
Assets	a compiler a website a service what people are working on expertise [research]results an email this is what we discussed this is what i am going to do databases software report	researchers printing new machine information systems my first big job The company tool some algorithm information the system ideas publications people things	personnel intelligence basics of how that was done employees freelancer Google engine e-mail systems open ERP system Wiki
Communication	Words relevant to the meaning of "communication" be in the same room be on the same page being accessible being afraid being open by myself come to me contact directly discussing email going to [as in making contact with] having differences in a group informal knowing making an appointment other people question receive refer to later send set in stone talk telling to meet to see someone Work Words and phrases expressing content of communication how to fullfill my job relevance that i should do something to handle something what i am doing what i am going to do what is going on what their expertise is What their involvement is	busy with apply for a job good ideas report time communicating with colleagues discuss with collaborate talk about what individual work discuss what discuss before discuss about some information discuss how a smart way to hundreds of ways better solutions important to discuss standard communication ideas speech email local collaboration talk meeting questions discussions depends on the task not the most social	e-mail communicate directly systems open ERP system Wiki
Structure	group moved to the whole our faculty university have seen it develop	central person department three sections MBSD section general support institute The company PHD students people within the project small group project group academics open to questions their own specific research institute the university divide three big sections reorganize three sections research institute education institute parallel structure practical organisation get paid to section combined group corporation group connections separate group doing ... for support director informal organisation individual work one big family	organization D currently expanding with more personnel company T 3 employees and 1 freelancer writing text working systematically organizing programs programing
Task	work on have been working on project [tasktype?] ov chip [name of project instance] participate in... research [tasktype] as a ....[fullfilling a role is a way of looking at having a task] researcher [something to be] programmer [something to be] part of my job to support [tasktype] maintaining websites [tasktype] maintaining services [tasktype] i should do i have to deal with daily work work for i have to fix it [task type] do work for stuff i do for my job what i am doing giving direction support question [task type] come to me [with it][which is used in this context as "giving me it as task"] handle it to ask [tasktype] to send an email [tasktype] to wait for [task type] to communicate back [tasktype]	currently busy with project involvement writing software tools for I do system administration general support solve problems have some problem solved some work done able to do my work individual work something I would like to do that we needed a smart way to hundreds of ways to do something implement it use the time for deadlines doing their own specific research programmer research is not my first get paid to do influence on my work research general things system administration it depends on the task have other things to do a lot of work to do have nothing to do else projects which are the backup experiment highest priority if there is a deadline, I don't have a choice individual work scientific programmer	full responsibility almost everything what to do targets of the customer linkbuilding writing texts improving the internal structure information gathering programming technical aspects
Goal	is about	intention reaching good ideas reaching a goal solve problems have some problem solved some work done writing as much papers as possible with good quality acquire knowledge enrich our society make quality of life better what you want to do what you should to do get better solutions deadlines nicest solutions specializing	web design online marketing customer gets more and more links and will be more visible in Google search engine most important are the customers, so their website is visible in Google web design of the customer
Hierarchy	formal boss making an appointment level consisting of group [name of a parttype] whole computing science [name of a part/whole instance] within institut [name of a part/whole type] above faculty [name of instance] university [name of an instance] higher [hierarchy] in [any organisation] giving direction [directing]	MBSD section (Model Based System Development) Frits Vaandrager organisation flat director the board of directors head of department not the professor the real authority in an university not really a hierarchy the most influence on my work my real boss it depends on the task education director flat informal see them as a director	owner of the company manager employees freelancer me on top and the rest two level
Processes	communicate a lot exchange results work together	busy with doing research job report time not a nine to five job steps. make sense better solutions came in later use the time for a meeting once a week totally specific different reorganize practical organisation better for the corporation if a lot of work to do if I have nothing to do else budgets cuts pressure will gain free to do your work as long as you deliver people work really hard	customer always comes to me how the work would be done e-mail webmasters of different sites to ask if they can place a link on their website to the site of the customer program Java applications give the person the whole task can have them work on the same task and still have no interference with each other. But that is a big thing I want to avoid because it will always make delay

discussion

When we look at the table we see that the content of the concepts is very different for the different interviewees. This could be due to the fact that the interviewees use different mental models and a different vocabulary to discus the topics but we suspect that part of the differences are attributable to the fact that the coding is done by different coders. One of the clues for a different way of coding is the fact that some of us have started out coding sentences or larger chunks of text and another has immediately coded words. This is visible when you look at the codes in the linked documents where you see the codes in the transcribed text. We all produced lists of words or phrases but still we expect the difference in approach to have had effects on these lists. Another clue is that when reading the content of the codes of others we found that there were elements in our transcript that we could have been coded under a certain concept but didn't because we didn't see the relevance.

All this said. it is still interesting to have look at the results and compare aspects of the lists that we think are comparable.

Assets

The lists for assets are relatively close to each other. They mainly contain examples of assets. we see that all three interviewees have examples of running software or information systems as assets. Also people or their capabilities feature in all three lists. There are differences. The only one that considered an organisation an asset was HK. It would be interesting to see if this was still true when the other coders had another look at their work, knowing this.

Communication

These lists are very different. One of them is very long and one of them very short. That is the first obvious differect. It could be the case that in one interview a lot was said on communication and in the other only a little.

Lets have look at what is in the lists. In both the RWS list and the HK list there are elements that are there because they refer to topics of communications. For instance "what is going on" (RWS) or "better solutions" (HK). In the JB list only means of communication are included.

Structure

Another example of very different lists. One of the issues may be that we have overlapping codes. "Structure" contains both "process" and "hierarchy". Some of the content that has been been coded under hierarchy in the RWS list would have been better placed in the Structure list. This would account for part of the difference between the RWS and HK list.

In the JB list there are two letters that stand for names of companies and so are instances of wholes. This only looks different but both of the other lists have references to instances of "organisation" so this is not really different.

A difference that could be attributable to the different model that the interviewee was using while talking is the fact that in the HK and RWS lists there is talk of groups and departments and in the JB list there is talk of people working for the company. This could be a reflection of the fact that the first two work for a very large organisation (the radboud university) and the other is leading a small company with only a few people working for it.

Task

In these lists elements that could be labelled as task type form the bulk of the list. This goes for all three lists. clearly, talking about tasks examples of tasks were abundant. It is not surprising that HK and RWS have overlapping sets of task types since they fulfill the same kind of functions in the same organization. It is also not surprising that the task types (or names of them) mentioned in the JB list are different since his work environment and his function is very different.

Another element that we see in all three list is the statement of some kind of relation to a task. "what i have to do", "a lot of work to do", "work on", "what to do".

Another thing that showed was that both RWS and HK seem to consider fulfilling a role to be the same as or asking to having a task.

Goal

Goal is represented both by the goal of an organization and goals of people working in organization. For example, JB, as a owner of a company, put organizational goals on the first place, and others consider their tasks as primary goals.

For people "web design" and "making customer happy" are goals. So tasks and the result are blended together in their vocabulary to represent the essential of their work.

In the coding of HK both words describing a goal and achieving a goal are coded. This is also because he discussed a more general goal of working at a university, compared to talking about the goal of his specific job. This makes sense because there is a big variety of positions at a university, compared to a marketing company for example.

Hierarchy

In all three lists for "hierarchy" there are words for boss. In the RWS list there is only one. It could be good to check if this is due to sloppy coding or really is a reflection of what was said in the conversation. The words used by HK and JB for boss are different and probably so because they work in a different environment.

Another thing shared by all three was the reference to levels: "flat", "higher", "two levels", "level", "above". We think it is safe to say that these people share "level" as a central element of the meaning of hierarchy.

Something that RWS and HK shared but JB not was the mention of formality as an aspect of hierarchy.

Processes

From the given lists it is not particularly clear whether we actually have words describing processes and that the words just happen to be similar to words using in tasks and communication, or that this similarity comes from the way we coded it.

We believe that processes are very close to tasks but also has elements of communication and goals. This is probably where the big difference between the list comes from (a more or a less strict distinction between the subjects by the coder). Maybe the category does not actually add any new information than combining other categories.

General

In general we can say that it could be a good idea to do another round of coding and also more explicit comments to relate a word or phrase to the code.

For example: in the RWS list for communication there is a phrase "be in the same room". This is under the code communication because in the conversation this was is considered something that is relevant for communication. "The connection of "be in the same room" to "communication" is very different than the connection of "talk" to "communication". To talk is an example of communication.

Another round of coding could have made the different lists better comparable. Implicit in the discussion of the lists for the concepts "task" ([tasktype] and [relation to task]) and "hierarchy" ([bos-underling], [formal-informal], [level]) is such a labelling and we can see that the comparison is much more fruitful there. It could also make differences in coding visible and make focussed checks possible. A new round of coding should also reduce the difference between actual instances (for example "writing many papers" as a goal) and more generic concepts (like "reaching" in goal).

Another general observation is that for most concepts the JB lists are smaller. Instead of attributing this to a different way of coding much of this could be attributed to the fact that the interview was shorter and part of the recording was lost.

During the discussion of the lists we could see that there are indeed some problems with intercoder reliability but also that for some concepts clear differences in wording were to seen and for instance for hierarchy it became very clear that the concepts of levels in the organisation was element for all three interviewees. This gives us an idea of what kind of results we could expect next time (with coding manual).