Computer-Mediated Cooperative Learning: Synchronous and Asynchronous
Communication Between Students Learning Nursing Diagnosis.
Ph.D. Thesis. © 1991 Dr. Rob Higgins


4.3 Findings

In this section the results of measures taken on the dependent variables are presented. These include counts of categorized items from the analysis of the transcripts (Subsections @Ref(ssfindcog) and @Ref(ssfindcoop)), ratings of the form and correctness of the nursing diagnoses and nursing care plans (Subsection @Ref(ssfindperf)), and responses to questionnaire items concerning impressions and reactions of the participants (Subsection @Ref(ssfindpart)). As well, in Subsection @Ref(ssfindpart), direct quotes from the interviews and from the open ended comments section of the questionnaires are reported.

In each subsection, the basic variables are listed with relevant excerpts from the transcripts or comments by the subjects to help clarify the system of categorization and evaluation. Next, the average weighted counts or ratings for each variable are presented in a bar graph depicting the difference between the asynchronous and synchronous groups. T-test statistics are also provided.

4.3.1 Cognitive Activity

Cognitive activity was measured by counting verbal elements in the transcripts which were categorized according to the criteria adapted from Powell (see Section @Ref(ssdaprepcode)). The following codes and categories were used:


	{MGM}	- managing the task
	{FRM}	- formulating problems
	{ARG}	- arguing
	{OPN}	- giving an opinion
	{CLR}	- clarifying
	{INF}	- giving information
	{QST}	- asking for information

(for further descriptions see the table on page @PageRef(tabconcod1))

This list presents the categories in a manner that begins with the higher-order cognitive activities reflecting self-management or executive processing functions ({MGM}, {FRM}) and ends with the basic information exchange activities ({INF}, {QST}). It should be noted that arguing {ARG} is a particularly intense cognitive activity since it requires integration and understanding of the opinions, issues, and problems under discussion. Further, as was discovered in the pilot study transcripts, in the face-to-face interaction, and through a preliminary overview of the core transcripts, arguing {ARG} was manifested in two distinct forms. The more specialized form of arguing involved direct responses to what was stated by one's partner and thereby suggested active debate. For these forms of arguing the {ARG} was appended with the code for responsive {RSP}.

As might be expected, there were many instances where precise categorization of a particular verbal element or sequence was difficult because of overlap among levels of content and the individual interactive style of the participants. For example, some verbal elements, though stated in the form of a question, are not simply asking for information. They are often problem formulation or arguing elements stated in a manner that invites a response. As such they were coded in terms of both the higher-order cognitive activity and the cooperative, facilitative activity. Nevertheless, if information was being requested (and not just suggested in an interrogative way), the verbal element was coded as a question regardless of the level of content (ie: task, problem, or case data).

In Figure @Ref(figexcerp1) a coded excerpt from one of the synchronous discussions is provided to demonstrate the application of the {MGM} code to elements of the discussion that reflect the effort to manage the task of establishing two nursing diagnoses. The more common {OPN} and {QST}, as well as the less common {PRS} (personal comments), are also shown. {FCL} will be discussed in the next section.


Figure 4-7: Synchronous Discussion Showing {MGM} Coding


n25:
{OPN}How about potential self harm related to poor
self-esteem?

n27:
{OPN}I like the poor self-esteem part. Actually it might
work. {MGM}We can keep this and come out with more. Right?
{OPN}How about, potential for malnutrition related to poor
eating habits.

n25:
{FCL}I was waiting for that one!
{OPN}Obviously this man needs a counsellor and the nurse
must be very careful to be positive with him and not
negative! {PRS}Did you take the computer course on
diagnosis?

n27:
{PRS}No.

n25:
{MGM}Let's just use those two for now and figure out a care
plan.  {QST}Which do you think is of highest importance?
{OPN}I think the potential for suicide is worse!


Figure 4-8: Asynchronous Discussion Showing {FRM} & {ARG}{RSP} Coding


n37:
   {FRM}For our nursing care plan, I think that we will be
able to achieve the diagnosis related to altered sleeping
and eating patterns more readily than we will be able to
improve patient's self-concept. {CLR}Maybe the second
diagnosis (altered sleeping and eating) should be the first
priority so that, once patient has achieved some sort of
health improvement through proper sleeping and eating,  he
can continue therapy to build up his confidence.
{FCL}That's all I can think of.

n35:
{FCL}{OPN}Counselling is a great idea. {OPN}Anti-depressants
would only prolong the real problem.
 {FRM}The order of diagnoses is important. {ARG}{RSP}If we
did work on the patient's self-concept first, his eating and
sleeping patterns may improve, since they may be directly
related. {CLR}Of course, proper eating and sleep patterns
may be a more immediate problem. {OPN}Maybe we should list
the second diagnosis (altered sleeping and eating) as the
first priority because seeing a therapist maybe a little too
much for the patient upon arrival at the hospital.

A further example of coding for cognitive activity is provided in Figure @Ref(figexcerp2). It is drawn from an asynchronous dyad in order not to misrepresent one mode over the other in these examples. This excerpt demonstrates the use of the {FRM} and {ARG}{RSP} codes. Compared to "managing the task", as shown in the previous excerpt (Figure @Ref(figexcerp1)), "problem formulation" is represented in verbal elements that deal with the issues of nursing diagnosis and nursing care planning at the conceptual and theoretical level rather than in terms of the task criteria (time and organization). The {ARG}{RSP} code is used for that part of the response by nurse 35 that addresses the ideas stated by nurse 37 in the previous message.

In order to reduce the effect of time factors, the total counts of verbal elements per dyad were weighted by a ratio of the word count per dyad divided by the average word count of all the dyads (see the COMPUTE statements in the SPSS-X program in Appendix @Ref). In this way, the volume of verbal interaction within each dyad helps adjust for variation in factors that may have affected the time needed to complete the task (eg: orientation to the equipment and environment of the study, and differences in keyboarding skills).

The graphs and statistics in Figures @Ref(fcogact1) and @Ref(fcogact2) show that synchronous dyads produced larger numbers of verbal elements demonstrating attention to managing of the task, problem formulation, and interactive arguing. Counts for these categories are emphasized because the differences between the groups resulted in at least p < .05 significance when the t-test was applied (using separate variances and one-tail probabilities). The synchronous group also produced larger numbers of verbal elements demonstrating attention to arguing (total), giving information, and asking for information, but the differences were less significant than those of the higher-order cognitive tasks mentioned previously. For "giving opinions" and "clarifying", the average for the asynchronous group was higher than that for the synchronous group by slight margins.


Figure 4-9: Cognitive Activity Counts (1)


Figure 4-10: Cognitive Activity Counts (2)

In fact, when the counts for all cognitive activity were combined, there was little difference between the two groups (see Figure @Ref(fcogact3)). Obviously, however, when the key cognitive activity counts where combined ({MGM}, {FRM}, and {ARG}{RSP}) the difference was considerable (and significant by the t-test (p < .001)).


Figure 4-11: Total and Key Cognitive Activity Counts

4.3.2 Cooperative Activity

Cooperative activity was measured by counting verbal elements in the transcripts which were categorized according to the criteria adapted from Beckwith (see Section @Ref(ssdaprepcode)). Two code categories were used: {FCL} - facilitative and {DBL} - debilitative (for further description, see the table on page @PageRef(tabconcod1)).

In Figure @Ref(figexcerp2) the {FCL} code is shown as it is applied independently to the expression, "That's all I can think of" since it acknowledges a limitation in the current thought process of one member of the dyad and thereby prompts the other member to provide assistance in the form of more ideas. {FCL} is also applied in conjunction with {OPN} for the expression, "Counselling is a great idea" since this is both the opinion of nurse 35 (that the idea is good), and an encouragement through the use of "great". Similarly, as noted in the previous section, questions that were used more for prompting one's partner than for requesting information were coded {FCL}{QST}.

Debilitative statements did not occur frequently. The average was less than one per dyad for both groups, and they did not occur at all in most of the discussions. When they did occur, they either protested the task at hand as in, "Do we have to do those retarded >as manifested by's Generally, in terms of cooperative activity, Figure @Ref(fcopact1) shows that the synchronous pairs demonstrated a larger portion of facilitative exchanges. These included efforts by the participants to make each other more comfortable, as well as clear statements of support, understanding, and encouragement.


Figure 4-12: Cooperative Activity Counts

4.3.3 Performance on Nursing Diagnosis Task

With respect to performance and outcomes, the nursing diagnoses and nursing care planning discussed by each dyad were assessed for form and correctness. They were also rated on the degree to which they adhered to the principles and guidelines used to establish nursing diagnoses and care plans. The variables are listed on page @PageRef(tagndvar) and the rating criteria appear in Table @Ref(tagndcrit) on page @PageRef(tagndcrit).

A good example of adherence to principles and guidelines is shown in Figure @Ref(figexcerp3). Whereas some asynchronous discussions also demonstrated good adherence (as seen in Figure @Ref(figexcerp2)), most diverged from the main task. In Figure @Ref(figexcerp4), for example, the pair begins to discuss nursing interventions in their first exchange without adequately formulating and stating nursing diagnoses.


Figure 4-13: Synchronous Discussion Showing Attention to Principles and Guidelines


n12:
Okay. Is depression classified as a nursing diagnosis?
n11:
I don't think so because it is a medical diag.  I can't
remember how we are supposed to state it.  Maybe self image
deficit.
n12:
How about hopelessness based on the things he said have
changed in his life and the way he sees himself.
n11:
Great! Shall we put it into their format?  Hopelessness
related to, as manifested by, etc.
n12:
Thats good, but I'm not sure exactly what to say its related
to. There are a lot of things it could be manifested by.
n11:
How about related to perceived failure? Or, I can't think of
a better word for perceived incompetence.
n12:
I like the perceived failure. It indicates how he feels and
can result from his parents high expectations.


Figure 4-14: Asynchronous Discussion Showing Divergence from the Task


n22:
Patient seems to be experiencing severe anxiety related to
his low self-worth.  Although his eating and sleeping
habits have deteriorated, I believe his depression should
be addressed first.  If we can get him to talk  more about
the circumstances surrounding his attacks, I believe we will
have a better chance at ameliorating his deteriorating
eating and sleeping habits.  Patient should be teamed up
with a psychiatrist and possibly a therapy group.
n26:
 I agree that his depression is a major priority and should
be looked after first. The only way to overcome the
depression would be for him to talk to someone about it.
The therapy would in all likelihood do him good. I suggest
that group therapy (after the initial one-on-one with the
therapist) where people with problems like his and some
that are worst (to give the perspective that there are
bigger problems than his) so that he has a wall he can
lean on for support.

Although separate variables were used to record the assessment of form and correctness of the nursing diagnoses and nursing care planning independent of one another, only their combined and averaged values are necessary to register the comparative outcomes of these dyad discussions. The graphs and statistics presented in Figure @Ref(fndncp1) show averages for the nursing diagnoses given first priority (ND1 = (NDF1+NDC1)/2), the nursing diagnoses given second priority (ND2 = (NDF2+NDC2)/2), and the nursing care planning (NCP = (NCPF+NCPC)/2). The two groups faired equally well in establishing the first nursing diagnosis, but the synchronous group did better on the second nursing diagnosis and on the nursing care plan. Similarly, as shown in Figure @Ref(ftotnd1) the synchronous dyads did better in terms of the total nursing diagnosis and care planning score (TOTNDNCP = (ND1+ND2+NCP)/3). This was true in regard to adherence to the principles and guidelines as well. Again, the t-test was used to demonstrate the inferential strength these differences.


Figure 4-15: Nursing Process Outcomes (1)


Figure 4-16: Nursing Process Outcomes (2)

The issue of the relationship between the cognitive/cooperative activities and the nursing diagnoses and care planning outcomes was not addressed in the focus questions for this study. However, the intuitively expected association was cause for further investigation. The scattergrams and statistics presented in Figure @Ref(fcorr1) demonstrate that key cognitive activity and adherence to principles and guidelines have a positive correlation with total nursing diagnosis and nursing care planning scores (p < .05, Pearson). Cooperative activity, as represented by verbal elements coded {FCL} for facilitation, did not show a significant correlation with these outcome scores.


Figure 4-17: Cognitive/Cooperative Activity Correlated with Nursing Process Scores

4.3.4 Participant Impressions and Reactions

Data on participant impressions and reactions to the use of computer-mediated communication for a dyad learning task were gathered from responses and comments on a post-task questionnaire, statements made in a debriefing interview, and personal comments contained in the transcripts of the online discussions.

The questionnaire was designed to elicit information about each subject's satisfaction or dissatisfaction with the nursing process content they developed, the cooperative interaction with their learning partner, and their use of the technology in general. It also provided space for individual comments (see Appendix @Ref(appposttask)).

The interview (see Appendix @Ref(appintschd)) was used to engage the subjects in a one-on-one, face-to-face situation in which they could discuss certain aspects of their experience. In particular, information concerning their use of the appropriate components and steps in establishing a nursing diagnosis was explored. Also, their opinion about the length of time provided for the task, the effect of interruptions or pausing in their discussions, and the relative intensity of their interaction at various points was recorded. They were also asked whether they thought the use of the computer to communicate was more or less difficult than dealing with the nursing issues covered in the discussions. Finally, they were asked to state what the best and worst aspects of the activity were.

In general, both groups of participants expressed great enthusiasm for the medium and for the opportunity to interact cooperatively with their classmates. There were a few minor complaints about the slowness of interaction due to either the medium or typing skills. Nevertheless, most subjects felt they had good interaction and that the medium was beneficial. In fact, the responses to the post-task questionnaire were so positive and consistent that additional statistical description would only be redundant.

The interviews were quite positive as well, but no patterns could be discerned from the subjects' comments on issues such as time, interruptions, pausing, and intensity of interaction. Most felt there was sufficient time (in either synchronous or asynchronous mode) and that the interruptions or need to pause in the online discussions were not a problem nor a benefit. In the same sense, it was not possible to pick out any pattern of intensity of interactive debate. Neither the subjects' responses nor the analysis of the transcripts revealed such a pattern, and certainly nothing indicated a difference related to the two modes. This was true in regard to "use of the computer" versus "understanding the diagnosis" also. That is, the participants were evenly divided on the issue and there was no pattern distinguishing synchronous users from asynchronous users.

The following are two samples of the comments written on questionnaires:


n35 - I felt scared at first, but by the second session I
      felt quite relaxed and enjoyed the work.  I feel I've
      learned a lot more about diagnoses/care plans than I
      did before, especially with all the "new angles" and
      thoughts my partner had.

n22 - I believe the experience might have been better had I
      been in direct contact (through computer) with my
      partner to exchange ideas because active communication
      seems to spur on more ideas.

From the interviews we have:


n17 (asynch) about the worst aspect:

About the whole thing?  Trying to come up with the nursing
diagnosis.  Sometimes you get stuck, and its like you're
sitting there, and its like, hmm, time's running out.


n5 (synch) about the best aspect:

I like the idea of being able to diagnosis with someone else
being there, I feel much more confident having someone to
assure me that my ideas or their ideas are right, umm, I don't
know, I just, I enjoyed doing this exercise and I found that it
helped me alot to work with someone else and accept their ideas
and realize that you know, maybe my diagnosis isn't right, and
maybe their's isn't and find some kind of, I don't know ...
[middle ground?] ... mmhmm.