Ethics is a set of moral principles that aim to prevent research participants from being harmed by the researcher and the research process. Ethical and moral responsibility is essential in any research, but when it comes to cross-cultural research it is even more important, as the researchers deal with individuals who have been exploited, who are more marginalised and vulnerable in so many ways. Often, they are people living in poverty, who do not have enough education to deal with the formality of research, and who feel too powerless to express their concerns or to resist the power of researchers.

How do we make contact with individuals and ask them to participate in our research, particularly if they do not wish to be found? Locating potential research participants can be a challenging and often problematic task. Many people are reluctant to enter the research field, as they do not ‘trust the researchers’, and may have other priority concerns (Liamputtong 2007a). Some groups, such as Hispanic Americans, have high mobility rates due to their employment and other issues and hence locating these groups can be a great challenge to researchers (Lange 2002). Because of this, recruiting research participants can be a formidable task for those carrying out cross-cultural research, and particularly so when the research is involved in sensitive issues.

In this chapter, I shall discuss several issues relevant to accessing potential research participants. I also provide some strategies which will assist researchers to gain access to and maintain relationships with the participants, so that their research projects may run successfully.

Cultural sensitivity is an important issue in conducting research with people from different cultures (Weinfurt & Maghaddam 2001; Papadopoulos & Lees 2002; Walsh-Tapiata 2003; Hall & Kulig 2004; dé Ishtar 2005a, 2005b; Birman 2006; Liamputtong 2008). Cultural sensitivity, according to Joan Sieber (1992: 20), is ‘the understanding and approaches that enable one to gain access to individuals in a given culture and to learn about their actual lifestyles (beliefs, habits, needs, fears and risks)’. In Phylis Eide and Carol Allen's terms (2005: 4), this is referred to as knowing the cultural context of the group that the researchers wish to work with. The researchers exhibit ‘cultural sensitivity and competence’ through their knowledge of the key values of the social groups. They also need to demonstrate ‘culturally appropriate communication and willingness to learn’.

Without appropriate cultural sensitivity, misunderstanding or (worse) racist attitudes may surface and this will jeopardise the progress of research, or at the extreme level, the termination of your project. Bahira Sherif (2001: 444) tells us about her experiences while conducting research in Egypt. Some of her American colleagues who dressed in ‘a very provocative manner’ repeatedly said that they were extremely displeased with ‘the unwanted attention’ from the men on the streets.

Introduction

Historically, cross-cultural research has been an important part of the anthropological discipline. Researchers within this discipline have worked with people in different social, cultural and geographical settings, using mainly ethnography as their method of data collection. They are known as ethnographers. They have tried to conduct their research with the hope that they can ‘interpret what is on the “inside”, through the voices of informants’ (Adler 2004: 107). This tradition continues. Although the ethnographers are performing cross-cultural research, in the past they have also been seen as the ‘takers and users’ who ‘exploit the hospitality and generosity of native people’ (Trask 1993: 7; see also Minh-Ha 1989, 2006).

This book comprises nine chapters. In the first chapter, I discuss the necessity of performing qualitative cross-cultural research. As in any good methodology textbook, I provide some theoretical standpoints that I believe sit neatly within the framework of cross-cultural research. Chapter 2 introduces crucial issues regarding ethical and moral perspectives in performing crosscultural research. Some general discussions on ethical and moral issues which have been debated in the literature, historical examples of research which have exploited many individuals and communities, and issues relating to ethical principles and informed consent are presented. This chapter also includes discussions on the risk and harm which may befall cross-cultural researchers themselves.

In Chapter 3, issues relevant to accessing potential research participants are raised. I point to some strategies which will assist researchers to gain access and to maintain relationships with their participants so that their research projects may run successfully. Chapter 4 discusses cultural sensitivity in cross-cultural research. I argue that cultural sensitivity is an important issue in conducting research with people from different cultures. This chapter provides some suggestions about how to become a responsible researcher when working with cultural groups.

The reasons I wrote this book are many. For one, I am the product of cross-cultural identity. My grandparents, apart from my paternal grandmother, were migrants who escaped poverty from the south of China and settled in the south of Thailand, where I was born and raised. Throughout my childhood, I was constantly made aware of my ‘alien’ status within the local Thai community. It was not only my ‘ethnicity’ but also my ‘poverty’ that continued to plague my childhood. We were misunderstood about so many things, and often people would look down on us – the alien and the poor family. I survived all of this and I have always vowed to myself that I would write something about cross-cultural issues when I had the chance, and that chance has arrived. This is the reason for the birth of this book.

Second, because of my own cultural identity, I have great interest in the lives of ‘cultural Others’ who are also marginalised in society. In particular, I have been touched by writers who come from non-Western societies or those who have been marginalised due to their race and ethnicity. The story that I found most touching was when the tennis star Arthur Ashe announced that he had AIDS, a People magazine reporter asked him: ‘Mr Ashe, I guess this must be the heaviest burden you have ever had to bear, isn't it?’ Ashe said: ‘It is a burden, all right. But AIDS isn’t the heaviest burden I have had to bear … Being black is the greatest burden I've had to bear’ (in Ashe & Rampersad 1993: 139).

A questionnaire is more than merely a collection of questions. We have chosen the questions we want to include in the survey instrument, we have chosen the appropriate tools to deliver those questions and accept the responses, and we have decided on an overall design or “look and feel” for the presentation of these items. Next, we want to combine the items into a single questionnaire and add the elements that make it a usable survey instrument. This chapter will cover the following topics:

• Customization, including skips or routing, randomization, and fills

• Flow and navigation, including question numbering and action buttons

• Error messages

• Help and instructions

• Progress indicators and survey length

The topics in this chapter relate to the interactive elements of Web surveys. That is, what happens when the respondent enters a value into a field, presses a button, clicks on a link, or performs some other action? They are also about progress or movement through the instrument. The way many of these tools are designed and implemented depends on whether a scrolling or paging design is employed and whether active content (such as dynamic HTML or JavaScript) is used.

I've made a distinction between verbal and visual elements (see Chapter 3), between task and style elements (see Chapter 4), and between dynamic and static elements (see Chapter 3), but there is one final distinction that can be made, namely, between visible elements and hidden elements.

In this chapter, I examine the basic tools available for constructing a survey question or, more specifically, for accepting the answer that respondents may provide. If we restrict ourselves to hypertext markup language (HTML) for now, there are only a limited set of input tool available to the designer. Despite this, I continue to be amazed at how often these tools are used with little apparent forethought. In the previous chapter, I noted that HTML forms are the dominant tools for delivering Web survey questions and alluded to some of the limitations of using HTML. Understanding each of the HTML form elements and how they are used is an important part of good Web survey design.

These input elements have both enabling and constraining functions. On the one hand, their shape and behavior suggest to the respondent the type of action that is permitted. If the input elements are used appropriately, and if respondents have any familiarity with completing forms on the Web, they will know what to do when presented with a certain type of input field.

On the other hand, the choice of input element often constrains Web respondents in ways not possible on paper. For example, a series of radio buttons allows a respondent to select one and only one of the available options. This is potentially both good and bad.

One of the challenges of writing a book such as this is that the Web is changing rapidly. The World Wide Web of today is not the same as the Web of a few years ago. It certainly will not be the same as the Web (or its successor) of a few years hence. Much of the debate about general Web design (and it applies equally to Web survey design) is based on different assumptions about bandwidth, browser capacity, and so on. At one extreme, we're told to design for the lowest common denominator. In other words, design for what the Web was. This could result in boring, plain HTML, with no interactivity and few visual or other enhancements. At the other extreme there are those who argue one should design for what the Web will be, requiring the latest versions of browsers and plug-ins and the use of high-end systems to bring all the interactivity and enhanced features to the Web experience. Somehow, we have to strike a balance between these two extremes, designing instruments and developing design guidelines that are useful both now and in the future.

My effort to find a comfortable middle ground means that I am writing for today's dominant technologies but trying to focus on general principles rather than specific details, with the intention that they should apply to the future Web as much as to the present.

In the previous chapter, I discussed the basic hypertext markup language (HTML) tools for constructing a Web questionnaire. In this chapter, I go beyond the basics in two ways. First, I look at extending basic Web survey design to include a variety of visual elements. While the presentation of images is a standard feature of HTML, the use of images in survey questionnaires has been relatively rare, so this can be viewed as extending the measurement capabilities of surveys on the Web. Second, I examine the many possible enhancements in HTML to facilitate a variety of survey measurement tasks. This includes a variety of interactive tools using client-side scripts and applets.

Images in Web Surveys

For many decades, the primary building blocks of surveys have been words, whether written on paper or spoken aloud by an interviewer. The advent of computer-assisted interviewing (CAI) did not change this primarily verbal focus, although the delivery methods were expanded from paper to computer screen, and from interviewer delivery to prerecorded delivery (audio-CASI). There have always been exceptions, but these were generally for specialized applications. Examples include smiley-face response options for low-literacy populations, show cards for words or pictures (pill cards, residential integration representations), ad testing and magazine readership surveys, feeling thermometers, visual analog scales, and the like (see Couper, 2005).

Once could think of the computer screen or Web browser as an artist's canvas – the choices one can make in arranging the various elements on the screen, the colors and fonts one uses to represent them, the choice of background color or image, and so on, are virtually limitless. Unlike the artist, though, the goal of a survey designer is not to create a work of art but rather to facilitate the task for which the instrument was designed. Furthermore, unlike the artist's canvas which, once completed remains relatively fixed and permanent – the artist has full control over the expression of his or her ideas – the design of a Web survey on a computer screen, as rendered by a browser, is subject to variations that are often not under the full control of the designer. Therefore, the survey designer does not have the free reign of expression open to the artist.

In addition, whereas the artist may not care that some people do not like their creation, and people may see the same piece of art in many different ways, the survey designer must of necessity care that the information content is viewed and interpreted the same way by a variety of different respondents.

Despite these and other differences between the two genres, it is surprising how many examples of Web surveys can be seen in which the designer's creative juices have apparently been given free reign.

Why a book on the Web survey design? This question really has three parts: (1) why Web, (2) why survey, and (3) why design? I will try to address the three parts of this question in this chapter, but first I will briefly describe the major types of Internet and Web surveys prevalent today. This will set the stage for the discussion to follow.

Internet and Web Surveys

In the relatively short time that the Internet, and particularly the World Wide Web, has reached widespread penetration in the United States and elsewhere in the world, Internet or Web surveys have rapidly emerged as a major form of data collection. It's sometimes hard to imagine that Telnet was developed in 1987, and the first graphical browser (NCSA Mosaic) was released as recently as 1992 (see www.ncsa.uiuc.edu), given the attention that Web surveys are getting in the survey profession and research literature. The adoption of online surveys has spread faster than any other similar innovation, fueled in part by the dot-com frenzy of the late 1990s but also driven by the promise of faster and cheaper data collection. For example, according to ESOMAR's global research study conducted in 2001, an estimated 20% of the U.S. survey research market could be conducted online by 2002, reaching 35% by 2004 (see www.esomar.org).