| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!

View
 

Method Requirements

Page history last edited by ted.coopman@... 12 years, 3 months ago

Data Collection Method Guidelines

Good research requires enough data to provide meaningful results. Depending on your research, here are what I consider the minimum amounts of data you need to collect for each method. No matter what you choose, you must have a solid rationale for why you chose to analyze what you did. PLEASE do not hesitate to have me look over any interview questions or surveys. I have included some general advice as well.

 

All choices for methods and choosing participants must have a solid rationale. That is, a detailed explanation for why that choice was made. For example, why did you interview or survey those particular people? Participants must have information that will help you address your topic/question. So, while you are on solid ground to use friends who are Facebook users to ask questions on how they use Facebook, interviewing your friend who works at Best Buy about the iPhone is not.

 

At least one method (data gathering) must be conducted primarily online. For example, conducting a survey via SurveyMonkey or Facebook, using Skype for your interviews, or collect data via the internet on organizations.

 

You must use a minimum of two methods, one qualitative and one quantitative. Some methods can be utilized as either qualitative and quantitative.

 

A Note on Dealing with People

Getting cooperation for participants in research studies can be incredibly challenging. This includes using teammates or co-workers, close friends, or even relatives. Moreover, even if you do get a pledge of cooperation, people still may flake on you - expect it. Here are some tips on gaining cooperation.

• Provide an incentive. The simplest inducement or incentive can create a degree of reciprocity and gain you cooperation. Offer to provide simple beverages or snacks for interviewees or focus groups (note: any "adult" beverages should be withheld until AFTER you are done), a raffle for survey participants may also help.

• Do not try and schedule participants during times they may be otherwise occupied such as breaks. midterms, or finals.

• Design your study to use people you know and have access to, so researching student behavior is always a safe bet.

• Use people with an built in incentive to help you. 

• Ask for cooperation, warn them, and remind them of their obligations.

• Choose people who have an interest in your topic.

 

 

Analysis of Websites or Online Media Content (qualitative or quantitative)

If you are planning to analyze online media content (games, programs, music, etc.) or websites it is best to either go deeply into one example (case study) or to do a comparative analysis between several examples. Please avoid analyzing TV or movies that are merely available online. Make sure to capture (download) any material you wish to analyze. Websites are dynamic and change often. Use archive features, screen shots, or print as pdf to save your data.

 

Here are few ways can analyze online content: 

 

Websites: There is really no end to the comparative analysis opportunities. You could compare news websites, political websites, auction sites, online dating services, airlines, etc.

 

Special Online Content: Video-Blogs, YouTube video, and other content made especially for online. If these are short segments (5-10 minute) then I expect you to review at least 5 segments. 

 

Video Games: A video game can be a great platform for research. Students have looked out how teams work in MMORPGs and ethical dilemmas presented in certain game scenarios.

 

Print-based online media: especially advertising is a rich source of data.

 

Minimum Requirements

• For individual websites as least 4 distinct sections (eg. Frontpage, About/Mission Statement, Members, several specific content section). In depth analysis of a single site will require more depth. For a comparative analysis between 2 or more websites, at lest 4 distinct and comparatively similar elements/pages.

• For serial video content (eg. Simpsons episodes) 4 individual full episodes for one, 2 episodes each if comparative (eg. Simpsons vs, Family Guy).

• For movies it depends on which elements you intend to examine and how. Ask the instructor for specifics.

 

For each additional research team member, you should double the data collected.

 

Things to watch out for >

The main problem for student researchers is that the analysis lacks depth or is too limited.

 

Interviewing (qualitative)

Interviewing can be complex, but it is also a great way to get at the motivations for people's media use preferences or perceptions of media. If you plan to use interviewing, first go to the Interviewing Workshop page and take the interviewing tutorial to find out the tricks and tips to make this method work for you. Please note that an interview schedule is not when you did your interview, but the list of questions that you have created to ask.

 

Citing your interview: interviews are data and do not need to be and should be cited like sources. Simply reference the interviewee inline when quoting or paraphrasing (eg. "Interviewee #3 stated that..." or "Roxanne explained...").

 

Poorly conducted interviews are painful and waste of time. There are two ways you can go about using interviews.

 

Minimum Requirements

• In depth Interview: If you plan to look deeply into individuals experiences, you need to go deep and have an interview schedule with at least 20 questions with 2 follow-up/probes each. You will need at least 2 distinct interviews. For our purposes this is not the best strategy unless you can get some time with someone important (A noted academic expert, CEO or other executive, etc.)

• Several Shorter Interviews: This is the preferred method for this project. You can get good comparative data by interviewed 5-6 people with interview schedules with 10 questions with 2 follow-up/probes each.

 

All interviewees should sign a consent form - Interviewee_ConsentForm_151i.doc - you must submit a consent form with your final assignment if you chose to interview people. Consent forms MUST be submitted as scanned documents OR as digital photos providing they can be easily read.

 

All interviews should be sound recorded unless interviewees specifically decline to be recorded.

 

For each additions team member, you should double the interviews required. For example 4 in-depth interviews for two team members and 6 for three.

 

Things to watch out for >

The main problems with interviews is scheduling and flaking. Interviews need to be schedule with large margins for errors and students should always have at least one back-up interviewee. The main problems for student researchers is that they fail to adequately code their interviews and neglect to quote their interviewees in the discussion.  Also, failure to have enough questions and required follow-ups.

 

Surveys (mainly quantitative, but can be qualitative)

 

Surveys can be a good way to gather more extensive, if more superficial, data. Facebook, Survey Monkey, and other online resources can get you access to a wider variety of people. A copy of the survey form should be included in your final submission. Here are some suggestions to guide you.

 

Keep it short. The willingness of people to fill out and return a survey drops off quickly as the length increases. I suggest 8 questions and definitely no more than 10.

 

Keep it simple. The more complex the question the more likely people will be unwilling to participate. Always test your survey on classmates or friends first.

 

Get used to disappointment. People you know are more likely to complete surveys for you, but in any event, if you get over 50% responding you are doing well. If you use Facebook to contact participants, list it as an Event. This will reduce you message as being seen as spam and will keep it foregrounded on the page. Send reminders regularly.  

 

Avoid relying on only yes/no and questions that simply generate numbers. A combination on numeric and short answers will yield the best results. For example, you could use a Likert Scale (look it up) to gauge how violent someone thinks a certain video game is, then follow that up by asking for an example, and perhaps how violence influences the attraction of that game.

Surveys are often helpful when paired with interviews. Use the survey to ask for interviewee volunteers.

Minimum Requirements

• You will need at least 40 completed and usable surveys.

 

For each additions team member, you should double the surveys required. For example, for 2 team members 80 

completed and usable surveys, for 3 team members 120. 

 

Things to watch out for >

Student love surveys going into projects and usually hate them coming out. It seems easy, but is actually a lot of work for often little pay-off. There are two main problems for student researcher using surveys. First, a good survey requires a lot of work upfront to craft good questions that may yield useful data. Often survey results are almost useless. Second, getting cooperation from participants is much more difficult than you would expect and often the quality of responses is very low. That is, you add in spaces for people to leave detailed explanations or descriptions but they do not. It is hard to get people to take the time to provide the information you need/desire.

 

Focus Groups

Focus groups (basically group interviews for our purposes) can be a great way to expose people to media/objects/situations and get their reactions. Often, groups can encourage discussion and exploration of a topic. However, it can be easy for a few people to dominate focus groups, so a strong moderator, a clear agenda, and formal topics/questions are needed to "focus" participants.

 

Focus Groups must have an agenda, plan, and a list of discussion topics and questions, and the demographic information for each member as well as a rationale for their participation, that need to be submitted with the final project.

 

Focus groups should always be recorded to capture all comments and discussion.

 

Focus group members should be identified by number, letter, or assigned name or other code that is linked to their demographic information.

 

Please avoid having an unfocused group discussion/meeting of friends and put off any alcohol consumption until AFTERWARD.

 

It is best to have a planned activity where participants review any materials, then answer questions on a provided form. This helps to focus participants, keep people from social loafing ("yeah, what he said"), and provided ready made recorded data.

 

It is recommended that one team member run the group, while another takes notes and operates any media. Switch off roles for different groups.

 

Focus Group members should sign this group release form (to be submitted with project). Consent forms MUST be submitted as scanned documents OR as digital photos providing they can be easily read.

 

Minimum Requirements

• You will need at least 8-10 people for your focus group.

 

For each additions team member, you should an additional focus group. For example, for 2 team members 2 groups of at least 8-10 people for your focus group, for 3 team members at least 3 focus groups of 8-10 people.

 

Things to watch out for >

The main problems for this method is "group think" in the sense of social loafing where participants simply agree with others' statements, the tendency to get off-topic and to perform roles for others in the group (especially for the other sex). If you are looking at gender differences or similar phenomena you may wish to consider have two or more smaller focus groups. Other problems for student researchers is that they fail to adequately code their focus group discussion and neglect to quote their participants in the discussion.

 

Semiotic/Visual Analysis

This type of analysis must be extensive enough to yield enough data for a complete analysis. The more sources you analyze, the less in depth you will be able to analyze. A comparative analysis of 2 or more sources (ads, programs, films, etc.) is recommended. Here are some examples to guide you.

 

If you choose to analyze a single website, then this analysis needs to be extensive and include multiple pages and sections within a particular site. If the website is regularly updated (eg. news websites or blogs) you should collect data on at least 3 different occasions.

 

If you chose to do a comparative analysis of two websites, then you should select several representative pages or sections for each. For example for two political party websites you could look at the home page, About Us, Donation, and perhaps several position/topic pages.

 

If you choose online media content, you want to be selective. Video and audio take quite a lot of time to analyze. For example, if you compare DIY series on YouTube, it would be a good idea to use similar themes (how to change your oil or car repair in general, gardening, etc.).

 

Make sure to capture (download) any material you wish to analyze. Websites are dynamic and change often. Use archive features, screen shots, or print as pdf to save your data.

 

Things to watch out for >

The main problems for student researchers is that the analysis lacks depth or is to limited and that symbols, colors, and other units of analysis are not clearly defined and overtly linked to specific meanings.

 

Observations

Observing people in an environment is both challenging and rewarding.  As with semiotics, this type of analysis must be extensive enough to yield enough data for a complete analysis. Any scene should be observed on at least 3 separate occasions.  This does not include preliminary observations (at least 2) that will prepare you to fine tune your research question, create coding categories, and collect data. You may even find that the venue you chose does not regularly have the behavior you wish to examine. Here are some examples to guide you.

 

1. Be very aware of the schedules and rhythms or your scene. For example, if you are examining participation in classroom, you will not get good data if you try to observe when a movie is shown, during an exam, or if class is canceled.

 

2. The more observation sessions you can arrange, the better results you will achieve. This is how you identify patterns.

 

3. It is usually better to observe extensively in one venue rather than superficially in several. Even if you choose to compare two scenes, you will still need at least 3 sessions each.

 

Things to watch out for >

The main problems for student researchers is not spending enough time observing or not doing good preliminary research. For example, intending on observing class discussion only to find that your class is watch a film or taking a test those days.

 

 

Comments (0)

You don't have permission to comment on this page.