Getting permission

If you're recording data from ordinary people, be sure they all give permission to be recorded in advance. (If you're using publicly available data, such as a reality show or TV or radio talk show, you can skip this step.) Explain that the recording is for your linguistics class and that your assignment is to document how people talk to each other (if you think this will change how they speak, you can tell them which class afterward). If you think you can do so without affecting the quality of the data, you can record your own interaction with one or two other people. Never record someone without getting their permission in advance. If you record a minor, you must have their parent or guardian's permission as well.

Recording spoken interactional data

You can use any kind of audio or video recorder that you have available; inexpensive audiocassette recorders are available in the bookstore. As a last resort, you may check out an audio recorder for 24 hours in the Linguistics Department Office (3607 South Hall). You'll need to supply your own analog cassette tape and batteries. If you use a Linguistics Department recorder, be sure either to make sure the external microphone plugged into the recorder is switched on or remove it and use the internal one built into the recorder. Also be sure to remove the external microphone so you can do playback. Make sure your recording equipment will pick up audio well enough that you'll be able to use your data. Always do a trial recording before you start collecting data! Check things like microphone range, tape length, problems with peripheral noise. TIP: Don't place the recorder on a surface that will vibrate (like a table); if you must do so, put something soft but not too absorbent, like a book or a pad of paper, under it. Check your batteries and have spare batteries and tapes on hand! Once you're ready, go ahead and begin your data collection. A reasonable goal is to record about 30 minutes to an hour of speech and activity; you'll end up using much less than this in your paper, but this will give you enough unselfconscious speech to analyze the interesting parts.

Where to collect data

Note: The following suggestions are not comprehensive; many other approaches are possible. Check with me or your TA if you have questions.

All of these situations can include speakers of all one gender and/or sexuality, or a mixed group. In sociocultural linguistics, all interactional data is interesting, so don't worry if the interaction you're analyzing seems "boring." Also bear in mind that synchronous online discourse (e.g., chatrooms) is also a legitimate source of interactional data. If you'd like, you can be a participant in the interaction you're analyzing, but only if you think you won't act unusual or otherwise dramatically change the results.

• an ordinary conversation about any topic (between friends, coworkers, intimate partners, family members, etc.)
• a classroom discussion
• interaction in a workplace (a store, a company, a restaurant, etc.)
• a meeting of a group or club (e.g., a religious group; a group with a shared ethnic background; a political group)
• a debate or argument (e.g., on television, on campus, in an online discussion group)
• activity-based interaction (language use among sport team or fraternity members, among children at play, between people engaged in any physical or mental activity that they'll let you record)

Some potentially interesting linguistic phenomena

Note: Again, the following suggestions are not comprehensive; many other phenomena may also be of interest. Check with me or your TA if you have questions. You should focus on either one linguistic phenomenon that occurs repeatedly in your data (e.g., interruption) or consider a cluster of phenomena that you consider to be related in some way (e.g., they're all part of Lakoff's "women's language").

• How people take turns: do they interrupt? invite someone to respond? etc.
• How long is each participant's turn?
• Uses of humor and joking
• How people flirt or talk about sex
• How people display and deal with disagreement
• Any one of the features identified by Lakoff as "women's language" (or "men's language"): do they seem to operate this way in your data?
• Address terms or referring terms: what terms do people use to/about each other, including people of other genders or sexualities (titles, nicknames, insult terms)?
• How people display power or solidarity
• Whether people use informal (slang, profanity, nonstandard forms) or formal/standard language
• Pronunciation and/or intonation
• Nonverbal communication: gesture, smiling, proxemics (how close people get to each other), eye gaze, laughter

Some ways of going about your analysis

• Demonstrate that women and men use some linguistic feature differently in your data
• Demonstrate that women and men use some linguistic feature in the same way in your data (this is more interesting than you might think, especially if it's a feature claimed to be used differently by gender)
• Demonstrate that women or men in a particular subgroup use language differently from women and men in other subgroups (e.g., Latino versus African American men; lesbians versus straight women, younger women versus older women) OR unexpectedly use it the same way in your data
• Discuss in detail how women or men in a particular subgroup use language in your data, without comparing them to any other group (you don't need to have a comparative component to have a valid analysis; remember, the features you find need not be unique to the group of speakers you're analyzing in order to be interesting).

Some ways to make your analysis good

• Don't overgeneralize: don't use the present tense or talk about all women and men based on your data.
• Don't focus on difference to the exclusion of other ways of viewing your data (e.g., similarity, or just an account of an interesting group or situation).
• Be as specific and explicit as possible; provide examples and analyze them in detail, using concepts from the class whenever relevant.
• For more information about writing up your analysis, see "Writing Tips for Sociocultural Linguistics."