Linguistics is a Science
As you begin to think about your research topic, keep in mind that you will be conducting a scientific study. Whether you are building a perception experiment or addressing a core grammar theory, there is a prescribed that must be followed.
Linguistic research can be quantitative or qualitative, depending on the nature of data you are gathering. In quantitative research, the researcher will gather precise numerical data with the goal
of supporting a hypothesis. The data are analyzed statistically and presented in graphs and charts.
On the other hand, qualitative researchers have an idea of what they are looking for, but this may change as the research develops. Subjects (or participants) are observed, interviewed, or asked to write descriptions of their experiences. The researcher will then interpret and present findings. Quantitative research tends to be more along the lines of the scientific method whereas qualitative research is more open to interpretation. The nature of your research project will determine which method is preferable. Moreover, it is possible to blend methods by, for example, quantifying qualitative data by assigning numeric values to different gradients of input.
Many universities require that you obtain certification for carrying out your research project. This usually involves an short online training session such as CITI. In some cases, you will be required to have each of your participants sign a consent form that you will create, based on the nature of your project.
The Elements of a Research Project
Selecting a Topic
Choosing a focused yet interesting subject is perhaps the most difficult task in the research process. I find that this is where my students tend to flounder. Remember, it is better to feel that your topic is too narrow rather than launch into a study that is too broad or vague. For instance, instead of looking at whether or not women are more polite than men, instead of looking at specific forms of politeness or impoliteness (such as women use ‘please’ and ‘thank you’ more than men do), focus on a precise social environment in which you expect to find them. And be concrete in providing a basis for your choice. Also, select a topic that you are comfortable with and interested in.
Regardless of which method of data gathering you decide on, there are certain elements that must be presented in your research project. First, you must carefully craft a clear and focused hypothesis or research question/prediction.
Remember that a hypothesis is not simply an observation, but a prediction that can be tested. For example, you may have observed that speakers of Ecuadorian Quichua use Spanish words. This is an observation that could be made by anyone who speaks Spanish. However, by adding variables that can be tested, you will be able to form a ‘working’ hypothesis, that can be somewhat general, such as, ‘the Spanish words used in Quichua seem to include mostly nouns, verbs, and adjectives’. A testable hypothesis will focus on perhaps when these words are used, or which part of speech is more commonly used.
The Literature Review
A literature review is not always necessary. However, if you plan on writing one, here are a few key concepts. First, keep in mind that you are not reviewing papers that have already been written on your topic. You carefully select and refer to the peer-reviewed publications/primary sources to build an argument for why your study is so important. Thus, you may begin with a very brief and general overview of what has been claimed concerning your topic. Then you can show how these have accounted for various types of data, but then you also show how the literature lacks current information on your topic, (the data that has not been accounted for, or material that has not not yet been written about).
Click here for more details on writing a literature review.
The methodology for scientific research is fairly straight forward.
First, you will need to decide on the source of your data. This could be a corpus of archived information, or participants from whom you will elicit data. Be sure that you will have access to enough data to test your hypothesis. In terms of participants, the fewer you have, the more stimuli (questions/tasks) you will need to create. The larger the number of participants, the fewer. (Of course, this is not always the case!!) Controls are needed to ensure that all sources of data share certain parameters, i.e., for participants: age, gender, etc., for corpora: dates of publication, type of speech (formal or casual) etc., so that your data is not skewed.
A reliable method for gathering data must be carefully thought out and formulated. Software programs now exist that can extract various types of linguistic information from text. If you are working with human beings, possibilities include surveys or questionnaires, recordings, or simply information compiled onto a document. Be sure to obtain the proper equipment. If you are observing a classroom of children, you will want to use a video camera with good sound so that you don’t miss any data. If you are interviewing your participants, or asking them to ‘do’ speech, use a tape recorder or Tascam. Don’t count on your memory to recall accurate information. Controls in this area include presenting all stimuli identically, i.e., on a computer screen, using a computerized voice, in the same location to control for noise and other variables.
You will want to verify that your participants are each able to perform the required research task. This can be accomplished by providing a ‘pre-test’ that mirrors the task without informing your participant about what you are looking for. If they know, they will be primed to give you exactly what they think you want. Or perhaps a simply questionnaire to make sure sure participants are not color-blind if a task requires color recognition.
After the data are gathered, they must be carefully organized in a way that facilitates interpretation. You may use excel sheets, or annotation software such as Praat or Elan.
Organizing and describing your results is key. Initial observations are presented in a narrative and should include statistics (tables, charts, etc.). This is a good time to discuss how variables interface, show outliers, etc. Label tables and figures clearly and provide explanations for selecting the tables and figures you use. Explain how they best demonstrate the phenomena you are presenting.
This section provides your readers with a more detailed discussion of your interpretation of findings, what they actually demonstrate, how they support the hypothesis, (or not), and what they contribute to the literature in the field. Before the conclusion, the writer should address the weaknesses of the study, what type of improvements could be made, as well as suggestions as to the next steps to further the study.
Everyone has their preference for citing and referencing literature, including your professor. The two most widely accepted styles for linguistics are:
Modern Language Association (MLA) uses brief citations (usually author(s) name(s) and page numbers) that are inserted into the text, rather than using footnotes. The references for these citations are listed at the end of the paper, and include all required information on each entry. Click here for examples.
American Psychological Association (APA) focuses on using an in-text citation that provides the author’s name and the year of the publication. A detailed list of these references are included at the end of the paper. Click here for examples.
Here is a great site to find various types of reference materials in linguistics.
Power Point Presentation
Here is an outline to guide you in creating a power point presentation.
- This is not always required.
- An abbreviated version should be used (approximately 100-150 words).
- Describe the linguistic element you are presenting.
- State the hypothesis.
- Explain why this topic is so interesting and needed. In other words, show how the current literature does not address your topic specifically.
- Include a literature review framing the need for your study.
I. Subjects and controls
- These depend on the source of data.
- Be sure to carefully control your testing environment.
- Methods for elicitation
- Verification of reliability (pre-test)
- State the facts about what you found.
- Organize your data by using graphs and tables.
- Significance of findings
- Restate your hypothesis
- Show support (not proof!) of hypothesis or lack thereof
- How you could improve on the research process
- What could be done as a next step in researching your topic
The research process is quite rewarding. Give yourself plenty of time by planning ahead and staying organized. Remember, your contribution to linguistics research could be revolutionary!