I used of a Lickert Scale of what people saw as cheap vs. expensive and quick vs.
slow and then link those Lickert scales to objective values. When making all of these graphs, often subjects gave responses that were a range of numbers for time or money.
For example, a subject might answer with “less than 20 minutes” or “2000-3000 NTD.”
In those cases, those numbers were averaged or defaulted to the numerical value given to get a single number to graph. If someone answered a time question with “a day” as a unit, it was coded as 12 hours or 720 minutes. The numerical value is not actually important, but assigning it allows it to appear on the graph and give some idea of the magnitude of the difference between that number and some of the other responses. Some people also didn't answer certain questions, so those are also left blank.
These graphs are for descriptive purposes only. With so few subjects, statistically significant trends cannot be inferred. However, they do help illuminate subjects where a change of perception might have occurred. The method still has merit, but it became abundantly clear that it would require a far greater sample size to produce meaningful results. In this section, I am going to omit most of Scaled Data that dealt with Perceptions of care in the United States, specifically the data about Expectations for the US and Frequency for the United States. After this thesis was defended the first time, I sent out a survey that just had the scaled questions to several dozen more people. I added ten more people to Group A and eight more to Group B. I cannot enter those answers as new data, but I did use them to check some of my initial conclusions.
First, let's examine questions related to Variable 2, Expectations. These are the responses to Taiwan Question 2.2 about the cost of treating a broken arm in Taiwan.
57
Table 2.
Here, the immediately obvious outliers are B2 and B9. Both of these people, when asked the question, gave answers in US dollars. When converted, this meant that their answers were orders of magnitude more expensive than some of the others. While this represents a loss of accuracy, it is interesting because it shows two areas where experience in the United States might have altered how they see cost. What's even more interesting about those two people, is that, despite being the most expensive values on the graph, neither of them perceived their expectation as expensive. This was a trend that held true in the additional survey data. The people who had lived in the United States had perceived similar dollar amounts as less expensive than those who had only lived in Taiwan.
These are the responses to Taiwan question 2.3 about their expectations of how long it would take to get treatment in Taiwan.
58
Table 3.
This graph is somewhat misleading at first. A1 is the obvious strange case. This subject said it would take a long time to see the doctor (a five on the Lickert scale) but when asked for an estimate responded with “three minutes.” This was her response to further inquiry,
“A1: Til the doctor appears?
Matthew: 對 (yes) A1: 三分鐘 (3 minutes) Matthew: huh?
A1: I cannot wait.”
This was an interesting way of putting herself into the mindset of someone who had a broken arm at exactly the time when they broke it. There is also B9 on the graph who provided a written response where the subject declined to provide a time estimate.
What is interesting about this graph is how much uniformity there is in the times themselves. Most subjects believed that they would get treatment within an hour, which speaks to the faith that people have in the Taiwan system. B7 is another interesting case.
59
He drew upon his direct experience spraining his ankle to answer the question. This was not an isolated event. Several subjects explicitly brought up past experiences when answering the question about Expectations.
These are the responses given when asked about direct experiences with the Taiwanese health system.
Table 4.
There are two subjects with abnormal data here. A4.1, who detailed a time her child got an eye exam and it was free. B2 talked about getting pneumonia and detailed a story about having private supplemental insurance that covered most of the cost. Without supplement insurance, he gave the cost as 2000 NTD but, in the telling of the story, the answer to the Lickert Scale question was forgotten and we moved on.
This data is not directly comparable to the data in Table 1. Table 1 had people reacting to a hypothetical broken arm while here, the question was for the “most recent experience” whether that was surgery or a flu shot. The scale on this graph has an upper bound of 6,000 NTD as a limit where Table 1 had an upper bound of 60,000 NTD. The cost and perception of cost is still surprisingly consistent given that some people went to the doctor for a cold while others did it for gastric surgery. No one perceived going to the doctor, no matter what they paid for it or what the experience was, as expensive. The biggest problem with the data for this research is that nobody in Group A ever paid more
60
than 500 NTD for a doctor's visit. This is a problem because it means that the study can't see how a Taiwanese person who's never been in the United States perceived a bill of more than 15 dollars US. Essentially, everyone had inexpensive care by Taiwan standards while I was hoping for an instance where someone from Taiwan had care that was
expensive by Taiwan standards but inexpensive by US standards so there could be a comparison.
As it stands, there are the interesting cases of B7 and B8 who paid over a thousand NTD and still thought their care was inexpensive, but their utility is limited because there were no members of Group A who did something similar. We could
contrast A1 to B3, B4, B5, B7, B9 and B10 since all of these people paid equal to or more than A1 and still perceived the cost as cheap, though that has to be contrasted with B11 who reacted the same way A1 did. The data is consistent with people from Group B having a higher threshold, but it is not conclusive.
In a future study, I would have asked this question with an addition of the severity of the reason they saw the doctor. The reason for seeing the doctor obviously makes an enormous difference when it comes to why people see the doctor and whether or not they perceive it as expensive. The reasonableness questions on the survey were an attempt to get at the same idea, but they were difficult to translate in practice and the answers inserted more confusion than they eliminated.
This is what the subjects had to say about the time to treatment of their direct experience.
61
Table 5.
Here, a few cases are notable. B9 answered the “actual time” question with “a couple of days.” Hers was a written survey sent to me from the United States so it was clear that the question had been too vague and she'd read it to mean. “How long would the treatment take?” This was a common mistake, though it was always clarified during the in-person interviews.
It's interesting that both A4.1 and A5 saw one hour as a long time to wait for treatment. Three people in Group B also had experiences where they waited an hour or longer. B2 waited a day in the emergency room with pneumonia but didn't see it as a long time while B8 had a health checkup that took two hours and he didn't see it as a long time either.
B10 waited an hour and perceived it as a long time; however, she was one of the members of Group B who lived in the United States but never used the system while she was there. So, going back to the Framework, she never made a loop around starting from
“Direct Experience with the United States system.” This would probably lessen the effect on her perceptions since she never was forced to make the contrast between the United States and Taiwan. Her experience also suggests that direct experience with the health system has more power to change perceptions than merely living abroad for a long time.
62
C3 and C2 both said very low wait times to see a doctor. C2 mentioned that he made an appointment first while C3 had contacts within the healthcare system and could go through channels. Or as C3 put it, when asked this question,
“[I] Just go in... sometimes they even wait in there for me. Make a telephone call, say I have some problems and they're like, "Please come in!" Sometimes, they do everything before I get there. What I'm saying is that not everybody gets this kind of treatment.”
C1, though, talked about the problem of the doctor she wanted to see being too popular, so there was no option for appointments beforehand and she had to wait for an hour in that particular instance.
This was a question where the expanded survey showed little difference. This would have been helped by the addition of a “severity” variable for the time visiting the actual doctor.
The responses to the three questions, “How often do you go to the doctor in Taiwan?”, “What does that mean in times per year?”, and “How would you rate your personal health?” showed a clear difference.
Table 6.
63
This graph is a bit cluttered but it shows evidence of a change, and it has
comparable pairs. There are some things that need to be pointed out. C3 listed a three for his health, but explained, “Now is 'very often,' because of age related stuff like
hypertension, blood sugar, these age related disease. At least a three.” His perception of frequency here is being explicitly altered by his standards for what is ‘often’ for someone of his age advanced age. He was the only one in the over 65 age category. The
distribution of health for the study was interesting with no one listing their health as being at all poor. All of the people, with the exception of C3, were going to the doctor less than the national average of fifteen times annually.
It also illustrates one of the most commonly listed largest problems of the Taiwan system; that healthy people are going to the doctor when they don't need to. A look at A4.1 and A4.2 reveals that both of them rated themselves as being healthy and both said they didn't think they went to the hospital very often; however, one of them went four times and the other went five times. Compare with B3 and B7, who went 5 and 4.5 times respectively. They went to the hospital at exactly the same frequency as A4.1 and 4.2 but both thought they were going to the hospital very often. This suggests that the perception of frequency is something that gets changed by living abroad. B2 expressed a common sentiment,
“Well, it's so inexpensive so that even the most minute illness, you feel like you need to make a visit to the doctor, and most of the time you can probably just stay off by taking better care of oneself and letting the immune system kick in rather than just going to the doctor immediately.”
From this graph, you can see that people in Group B had a lower threshold for what counted as “often” when it came to going to the doctor.
The question about Taiwan's convenience and quality were the most surprising of all the scaled questions.
64
Table 7.
This result flew entirely in the face of expectations. Compared to the US system, Taiwan is astoundingly convenient, with many people listing that as one of the biggest problems with the system. This question expected the members of Group A to feel that Taiwan was inconvenient at a higher rate than Group B but the reverse seems to be the case. The quality issue is interesting to tease out. B1, B2, B3, B5, B8, B10, and B11 all gave the Taiwan system less than Excellent Quality. B1, B2, B7, and B11 all listed the quality of American doctors as the best part of the US system.
B5 is unexpected, however, since he seems to be operating in exactly the opposite direction that was expected. The explanation is likely due to who the subject was. He had recently arrived in Taiwan from the US and it was the first time he had been insured independently of his parents. When asked about choosing his doctor in the US, he mentioned that his parents chose for him, and when asked about his direct experience with doctors in the US, he remembered paying, “it was like 10 or 20 dollars. So I guess 20 dollars isn't very cheap, but it's cheap enough.”
The scaled answers to the same question about the United States are helpful here
65
and help explain some of the perceptions for the Taiwan system.
Table 8.
One of the things that quickly becomes apparent is how varied the experience was. The blank spots in the graph are because some respondents didn't use the medical system while they were in the United States so they didn't respond to these two questions.
For Group C, C2 and C3 did have direct experience in the United States, but it was too long ago to be included. With the exception of B4, B5 and B10, all subjects found the convenience of the American system to be equal to or worse than the Taiwanese system.
Only two of them B5 and B11 found the quality of the American system to be better. For all of the expense of the United States system, most of the Taiwanese subjects saw no corresponding increase in quality.
66