Week 5 Research Journal

Creswell: Chapter 12 "Survey Designs"

(focus on instrument design and ethics)

Before creating your own survey instruments, you should check for what is available. You may be able to find something that will work "as is" or something that can be modified to fit your needs. 

Steps for designing data collection instruments:

1.    Write different types of questions

2.    Use strategies for good question construction

3.    Perform a pilot test of the questions

Types of questions: background/demographic questions, attitude/opinion questions, and behavior questions

Sensitive questions deal with topics that must be treated with great tact; only use these types of questions when necessary for your study

Closed-ended Questions: allow for easy comparison, but not very deep answers; these can be beneficial when creating sensitive questions because the close-ended nature gives respondents comfort from knowing the answer parameters

Open-ended Questions: yield in-depth answers; places no restraints on the respondent; more difficult to code and analyze

Semi-closed-ended Questions: the best of both worlds; ask a closed-ended question, but allow respondents to also have the option to provide their own custom response

**Constructing good questions is the key to getting good results! "Good questions are clear and unambiguous, and they do not confuse the participants."

Possible problems with questions: question is unclear; multiple questions; wordy; negatively worded; includes jargon; includes overlapping responses; unbalanced response options; mismatch between the question and the answers; includes overly technical language; no all questions are applicable to all participants

You can spot any potential problems with questions by doing a pilot study and asking participants to identify any problems (i.e. poorly worded questions, responses that are confusing, and/or questions that take too much time to complete).

Interviews have higher response rates due to prior consent; questionnaire responses can be spotty. To get higher return rates, you can pre-notify participants, then mail the questionnaire, then follow up twice (two weeks between) to encourage nonrespondents to complete the questionnaire. The whole process should take approx. six weeks.

Response bias: responses do not accurately reflect the views of the sample or population.

Wave analysis: checking responses at specific intervals to see if bias is displayed; check it against the previous interval to see if there has been a big shift.

Ethical Issues in Survey Research

As with all studies involving human respondents, the main ethical concern of the researcher should be to keep the respondents safe. If the researcher uses interviewers to go out in the field, then the researcher is also responsible for the safety of the interviewer. Confidentiality should also be ensured by not sharing respondent information outside the research team, only identifying respondents by an ID number, and not accidentally exposing participants based on ultra-specific subsets in the data.

Langley and Meziani: "Making Interviews Meaningful"

(focus on interviews)

Interviews are a staple of qualitative research and many think they are simple; however, this article proves that they are more complex than many people realize. Interviews can be divided into five genres.

1. The Investigative Genre: seeking descriptive chronological accounts for an accurate view of the info; focus on events

2. The Apprentice Genre: the interviewee is imparting knowledge on the interviewer (the apprentice in this case); verbalizing thoughts while performing a well-defined task, as an example; focus on tacit knowledge

3. The Interpretive Genre: inspired by phenomenological thought, this genre captures the interviewees lived experience through the interviewer promoting openness and trust; focus on meanings

4. The Discursive Genre: life stories, self-portraits, narratives of practice, autobiographies, collective life histories; can use multiple interviews to complete the whole story; focus on communication practices

5. The Interventionist Genre: action oriented; dialogic inquiry; appreciative inquiry; interviews are tools for intervention; focus on reflexivity

Gawlik: Chapter 5 "Focus Group Interviews"

(focus on focus groups)

What is a focus group interview? Research conducted in an interview format (moderated by the researcher) which is held in a group setting and is freely focused on the main theme of the research.

By conducting FGI, you gain more in-depth insight into your topic without the pressure of having to follow specific questions. There is a basic scenario that is guided by the moderator, but the participants also guide the discussion, meaning they experience more autonomy within this type of research approach. Moderators try to incorporate open-ended questions to facilitate discussion. This allows for deeper exploration of particular topics that seem of particular interest to the group.

The group setting results in "group stimulation," which results in 1) increases in respondent engagement, 2) a sense of group safety, and 3) a domino effect of sorts where the participants not only respond to the moderator's questions, but also to the answers given by other group members.

While this may seem great at first, this can also lead to "group think," where respondents agree with responses from other group members, even though they may actually feel a different way. Therefore, we must guard against unanimity in the FGI setting.

Criteria for selecting participants must be well defined and contain appropriate sampling measures. Additionally, the scenario (planned sequence of discussion) is also key to setting the tone of the interview. This documents needs to include any majorly important questions which need to be discussed, as well as a time frame for each question to help the interviewer know when to nudge the participants towards the next question. This document is usually composed of key words and phrases to help keep the interviewer organized, but not tied down. In addition to controlling the pacing of the interview, the moderator also decides which potential topics introduced by the participants should be explored more.

Focus studios seem a bit creepy to me. While the recording equipment isn't necessarily hidden, it's also slightly hidden so as to not pull focus. These rooms can also have TVs, flipcharts, DVD players, a projector, something like a cork board to display visual aids, and it should be decorated in the homelike manner to make the participants feel relaxed. While none of this is particularly creepy, the inclusion of a one-way mirror which marks the observation room really upsets me. I know that there is nothing untoward about this setup in an ethically sound interview, but I've watched too many shows where bad things happen because of one-way mirrors to be fully comfortable with this setup. However, that's my own bias.

Types of Focus Group Interviews

Brief groups: discussions are brief and limited; doesn't give much time for actual discussion of topic

Extended groups: at least three to four hours; requires the comfort needs to the participants be met; time needed to breaks, meals, etc.

Mini-groups: smaller groups which increases ability to concentrate and cooperate; may be used when very specific populations are needed for a study (or when finances are tight)

Affinity groups: members are acquainted with each other some way; removed awkwardness about talking around strangers and participants are more willing to discuss personal topics in greater detail if they are already comfortable around their fellow participants

Class groups: picking groups based on their stance on the topic being studied with half the participants supporting one side and half the other; takes more of the form of a debate

Creative groups: participants are selected based on their special creative potential; used frequently when seeking to brainstorm for things like new products and new packaging

 

Discussion Question #1: How do we know that the responses aren't reflecting the views of the sample? The text says to watch for overwhelmingly positive or negative responses, but what if that's just how everyone feels?

Discussion Question #2: Do participants need to be made aware if the observation room is going to be used? Or do they just get briefed about it after the interview? 

Discussion Question #3: Are there particular types of studies that are more aligned with FGI? With regular interviews? With questionnaires? How do we decide which is the best method of data collection for our study?