Content analysis

The idea of the second study was triggered by an interesting observation that the customers pay attention to different hotel features in budget hotels, very different from their behavior in full service or luxury hotels. Since by the time of this study, there was limited understanding on the micro level such as customer satisfaction within the budget hotel sector. So we started this study via examining the online comments or reviews–sometimes we call it reviews on Home Inns. Nowadays people like to share comments and reviews on the internet, and this has become a very good source for us to understand the customers perception about budget hotels. So we extracted 3000 comments from the official website of Home Inns, and did content analysis.

We used a software called Nvivo.

This is pretty much how the raw data looked like. So the data is mostly in Chinese language. As budget hotels in China mainly attract domestic customers. Some of the comments are very brief， only a few words, a few characters, but some quite elaborate. But even with brief comments, they come with a good amount of information. Often one sentence includes a few million units or text units, as we called it in the paper. With this kind of context, the main objectives of the study are to identify what aspects of budget hotels that customers focus on in their online comments or reviews, and what factors contribute to their satisfaction or the dissatisfaction of budget hotel guests in China.

So with this understanding, we would be able to provide suggestions for budget hotel managers to improve their quality of product and service. So the main analyzing technique we used is content analysis. And as you can see, there are many, many items for us to analyze and record the result of the analysis. So it is difficult for us to manually do the work. So we used a software called Nvivo, which is very useful. This software can help us organize and record our codings and categories in a very systematic way. And frequency counting becomes much easier. Reading of the content under each code also becomes easier. So for this analysis, we use the text units as our analysis basis.

As I said just now for the content analysis, we examined the comments by text units. For example, for a comment like this, the comment says “The price is good. The room is clean and tidy. The overall environment is quiet, but the breakfast is not so good”. So we break it down to four text units. The first one is “the price is good”, and we coded it as “price”, and we give it a positive label because this is a good comment. And the second code is “the room is clean and tidy”, we coded it as “cleanness and tightness of the room”. And also this is a positive comment.

And the third text unit is the “overall environment is quiet”, we coded it as “quiet environment”. And the last one is negative. It’s about breakfast. So the coding is “breakfast” and it’s negative. So you can see that these codes are very specific. And the number of code gets big. So we categorize these codes into categories or themes as we sometimes call it. This figure shows the broad categories. And under each category there are different items and codes. We manually analyzed 200 randomly selected comments among the three of us. And we discussed on the categories and possible coding. We have debates until we reached an agreement, we created a framework for future analysis.

So with this framework, the three of us work separately using the software Nvivo, which is the you know software to sort out qualitative data in an efficient way as I said just now. These two tables show some of the details and results.

So as you can see from this table that each theme is broken down into different sub themes. All together, we have one, two, three, four, five, six, seven, eight categories. Each category, for example, “guest room” is divided down into these sub categories. So with each sub category, we have the total number of text units. And we have percentage, we have positive comments, and number of positive comments, and number of negative comments.

And there are a lot other tables in the paper. In the third study, we tried to develop a measurement scale on cultural values and behavioral norms of the Chinese mass travelers. Since the items for the scale was not available, the first step was to read the literature

and to interview the travelers: what do they value and how do they behave when they patronize a budget hotel? So in other words, the first half of the research is done in a qualitative way. And the interviews have helped generate these items. Some of these are in line with literature, some of them not. So with this list, we were able to continue the next step in the study, which was to design a questionnaire and collect quantitative data. So a qualitative study still very important, as you know in this type of study as a foundation of understanding the nature the study.

The above cases show the importance of qualitative research approach, although up to this time point, the opportunity of publishing pure qualitative research is slimmer than the quantitative counterpart. On the other hand, the mixed method approach is seeing an increasing trend, as you can see from this table. A mixed method is coming up. Quantitative obviously you know has more chance to get published than the qualitative ones. But the importance of this method cannot be neglected. These are the reference that I have used in this lecture. Thank you very much.