Introduction
The overall purpose of this assignment is to have students thoughtfully engage in the collection, compilation, analysis, interpretation, and presentation of 'real-world' data using Microsoft Excel. Due to the time-consuming nature and complexities of data collection, it is recommended that this be a group assignment.

This assignment can be tailored to a number of learning communities across the various concentrations and disciplines, as there are a number of different options for data collection.

Learning Objectives

• Understand the difference between quantitative and qualitative research
• Understand the strengths and weaknesses of (____insert research model here___)
• Clearly define constructs and operationalize them
• Understand the role of logical inference
• Appreciate the complexities of the research process, from data collection to data interpretation
• Use data effectively to support claims
• Present data summaries effectively
• Address strengths and weaknesses of data

Competencies Addressed
Critical thinking, problem-solving, communication, and group interaction.

I. Conceptualization and Design
The first phase of this particular assignment is for students to generate key questions, hypotheses, and/or constructs of interest. Small groups can brainstorm their key questions, hypotheses, and constructs of interest either in class or in synchronous (e.g., MOO) and/or asynchronous web-based environments (e.g., TownHall).

II. Data Collection
The data collection phase of this assignment can occur using a number of different methods. You might choose to have students administer a survey, conduct an interview, engage in observations, carry out a content analysis, or use a combination of these methods.

Examples of each of these different data collection strategies include interviews, observations, content analyses, and surveys. Although different concerns emerge when using the various strategies, you may want to devote at least one class period to discussing these various issues and your expectations of the students.

Remember, in each data collection strategy, students are gathering information for the purposes of exploring an area of interest, examining a particular theory(ies), or testing a specific hypothesis(es). Before they set out to do any kind of data collection, they must have a set of research questions and/or hypotheses.

[NOTE: Please refer to supporting materials (and potential handouts for students): Considerations when Choosing a Data Collection Strategy and Advantages and Disadvantages of Data Collection Strategies.]

III. Data Analysis and Interpretation
Once students have collected their data, they will need to analyze and interpret it critically and meaningfully. Please refer to the handout called Basics of Data Summary: Definitions and Using Excel.

Depending on your expectations, you will want to allocate a reasonable portion of in class time to discuss:

• The importance of telling a story with data - what are the key themes?
• Strategies for summarizing data - which of the quantitative pieces is it most important to tell your audience about?
  
  • Using frequencies and percentages (number of actual responses in a specific category)
  • Using measures of central tendency (mean, median, mode)
  • Using measures of dispersion (range, standard deviation)
• Good practices for visual representations of data (e.g., bar charts, line graphs, pie charts, histograms, other tables, etc.)
• Strategies for including qualitative data - what kinds of qualitative information do they have? Can they identify themes? You may also articulate the value of using example quotations (or other evidence) to supplement their quantitative information.
• Advantages and disadvantages of the method they chose. [Refer to Handout: Advantages and Disadvantages of Data Collection Strategies.]
  • What are the strengths of their work? The limitations of their data?
  • What conclusions can they draw with confidence?
  • What unexpected results did they encounter?
  • Are there possible alternative explanations for their findings?
  • What suggestions or considerations do you have for future investigations?

NOTE: If faculty feel comfortable and want to cover such ideas in class, more sophisticated analyses in Excel can be run, such as t-tests, correlations, regression, ANOVA, etc.

IV. Final Project Summary
There are a number of options for a final project summary. If you have had students working in groups, then you might choose to have students deliver some kind of group presentation either in class or on-line. Alternatively, you might have student groups post group papers on-line. There are various technologies that might be used to facilitate this (e.g., PowerPoint, a webpage, TownHall, etc.).

V. Reflective Component

For students
The final part of this assignment asks students to reflect on their learning as a function of this assignment. This might be incorporated into their final product in some way, it might be a separate reflective paper, or it might be part of the final course portfolio. Whatever you choose, ask students to consider:

• what was easy, hard, and/or unexpected about this assignment,
• how this assignment contributed to their development in the competencies outlined as part of the assignment
• the extent to which they believe they met the assignment's learning objectives
• what they would do differently if they had to complete this assignment in the future.

For faculty
As you facilitate class discussions, answer questions, and grade student work for this assignment, consider keeping your own list or journal of learning lessons. Possible questions to consider include:

• What did you feel you might have done a better job articulating? How might you change your presentation/discussion of the assignment or material?
• What student questions continually surfaced? That is, what issue(s) did they seem to have an especially hard time grasping?
• As you consider their end product, what did students seem to do especially well? With what were you dissatisfied?
• How much time did you allocate in class for this assignment? Was it too much, not enough, or just about right?
• What unexpected learning took place in the context of this assignment, either as you consider what your students produced and learned or as you reflect on your own expectations?
• If you used this assignment in the future, how might you change it to more effectively meet your needs and/or learning community learning objectives?

VI. Suggested Elements of Final Project/Presentation

• Introduction to the project
• Research questions, hypotheses, and/or key constructs of interest
• Summary of key decisions in data collection (e.g., operationalizations of key constructs, description of method(s) used, example data collection forms, etc.)
• Group summary (analysis) of compiled quantitative and qualitative information. This might include any tables, figures, graphs, or charts that they created, as well as whatever records or evidence of their raw data that they have kept.
• Summary of group knowledge (interpretation). This should include some discussion of strengths, weaknesses, future directions, etc.
• Group and/or individual reflective piece

::Advantages and Disadvantages of Data Collection Strategies

Structured Interviews

Advantages:

• Treats each participant as unique individual, builds on their reality and experience.
• Fairly standardized format allows for comparison of responses across individuals.
• Can ask questions in a number of formats, ensures that participants understand the meaning of the question.
• Can take advantage of opportunities to learn that present themselves that may have been unforeseen.
• Adaptability/ flexibility/ can expand on developing ideas, in comparison to surveys, can clarify answers (probe), encourage respondent to elaborate, get fuller answers to open-ended questions.
• Can control sequencing of questions (e.g., skip questions, etc.).
• Response rate can be potentially increased.
• Permits observation of non-verbal cues.

Disadvantages:

• Interviewer bias: attitude, tone, body language may affect the participant's responses.
• Interviewing is time-consuming.
• Open-ended questions are not easily reduced to numbers.
• Responses may not be accurate, honest, or there may be memory recall problems.
• It is difficult to compare across cases if not all the questions were the same.
• Pre-existing biases may have affected the questions asked and interpretations given.
• Depending on design, the data may not be easily reduced to numbers.

Observations

Advantages:

• Shows us how people behave in everyday life.
• Can interpret individuals' behavior in the context of their environment.
• Can do systematic counting along with qualitative observations.

Disadvantages:

• Cannot establish cause and effect, difficult to pinpoint the direct cause of the behavior or the exact meaning of the counts.
• Observer presence may cause people to act differently than they would normally behave.
• Observer bias: Individuals have a tendency to see what they want or expect to see.
• Not likely to observe undesirable or infrequent behaviors.
• Observations are time-consuming, particularly if you deal with reliability concerns through observer training.
• Depending on design, data may or may not be easily reduced to numbers.

Content Analysis

Advantages:

• Can do systematic counting along with qualitative observations.
• Typically inexpensive in terms of time and money.
• Allows for study of processes occurring over longs period of time.
• Method is unobtrusive, seldom having any effect on the subject being studied.

Disadvantages:

• Cannot establish cause and effect, difficult to pinpoint the direct cause of the behavior or the exact meaning of the counts.
• Reliant on recorded communications.
• Can be difficult to translate existing records (e.g., magazines, TV shows, statistical records, etc.) into quantifiable indices of some general concepts.

Surveys

Advantages:

• Less time intensive, can get a lot of information in a short amount of time.
• Surveys are relatively inexpensive.
• There is a standardized format, can directly compare information across individuals.
• Respondents may have greater confidence in anonymity and thus be likely to respond more candidly.

Disadvantages:

• Responses may not be accurate, honest, or there may be memory recall problems.
• Need to be careful about wording of questions, e.g., accessible to targeted age group, educational background, class backgrounds, or ethnic groups.
• Experimenter bias: in wording of questions or attitude and behavior of the surveyor.
• Survey information generally does not penetrate very deeply below the surface of an issue.
• Reliance on written/ reading comprehension.

In each data collection strategy, students are gathering information for the purposes of exploring an area of interest, examining a particular theory(ies), or testing a specific hypothesis(es). Before they set out to do any kind of data collection, they must have a set of research questions and/or hypotheses.

Interview data
A primary task here is for students to generate a set of interview questions. Issues to explore with students include:

• The choice to use open-ended questions, close-ended questions, or some combination of both.
• Guidelines for writing questions or items and length considerations.
• Demographic information. What groups might they want to compare across and why? What do they need to know about their participants?
• Sampling, response rate, and generalizability. For example, whom do you ask to fill out your questionnaire and why? What happens when only a small number of individuals agree? How generalizable are your findings?
• Research ethics, including confidentiality and anonymity
• How will they design their interview form in order to increase the reliability of their data collection efforts?
• Good practices for conducting interviews, emphasizing interviews as a social interaction, e.g, obtaining consent, professionalism, importance of eye contact and body language, listening and clarification skills, appropriate dress, etc.

Observational data
A primary task here is for students to decide what behaviors to observe and how they will specifically define (or operationalize) these behaviors. Issues to explore with students include:

• Defining their constructs of interest in measurable, specific terms in order to get some quantitative counts
• At the same time, it is important to remain open to novel events and insights - what kinds of qualitative things do they feel are important to record?
• Where will they do their observations and for how long? Are there any strategic considerations here given their theories and interests?
• Research ethics, including confidentiality and anonymity.
• How will they design an observation coding or rating sheet in order to increase the reliability of their data collection efforts?

Content analysis data
A primary task here is to determine what concepts are of interest to them and how they will specifically define (or operationalize) these concepts. Issues to explore with students include:

• What materials to use as their sample, e.g., women's magazines, television advertisements, billboards, MTV videos, commercial websites, or other specifically defined public spaces.
• The parameters for data collection. For example, how many magazines will each person code? How many hours of television or how many music videos will each person view? Does the rating of a television program matter? Is the time of day an important issue to consider? Will individuals attempt to code the same material or will they intentionally target different material?
• What specific information or categories will they code within their sample of materials?
• How will they design a coding or rating sheet in order to increase the reliability of their data collection efforts?
• Research ethics, including confidentiality and anonymity

Survey data
A primary task here is for students to generate a set of survey questions. Issues to explore with students include:

• The choice to use open-ended questions, close-ended questions, or some combination of both.
• Guidelines for writing questions or items and length considerations.
• Demographic information. What groups might they want to compare across and why? What do they need to know about their participants?
• Sampling, response rate, and generalizability. For example, whom do you ask to fill out your questionnaire and why? What happens when only a small number of individuals agree? How generalizable are your findings?
• Research ethics, including confidentiality and anonymity · How will they design their survey instrument in order to increase the reliability of their data collection efforts?
• Good practices for survey administration, e.g., obtaining consent, professionalism, appropriate dress, etc.
• (If faculty feel comfortable) Creating scales and more detailed measurement concerns.

::Basics of Data Summary: Definitions and Using Excel

Strategies for Summarizing Data

Frequencies and Percentages

Frequency:
The number of actual responses in a specific category; how often a response occurs.

Percentage:
The number of actual responses in a specific category divided by the total number of responses; percentages are essentially relative frequencies.

Measures of Central Tendency: look at the most typical score in a data set

Mean:
The average score (sum of scores divided by the total number)

Median:
The middle score (organize scores from lowest to highest and find the middle score)

Mode:
The score (or response category) that received the most responses (organize scores in a frequency distribution and find the most common score)

Measures of Dispersion: degree of differences, or variability, within a data set.

Range:
Defines the spread of the scores from lowest (minimum) to highest (maximum).

Standard Deviation:
Roughly the average distance of scores from the mean.

• The standard deviation (SD) also gives you information about percentiles. Assuming a normal distribution (a bell curve), 68% of scores fall within one SD of the mean. 95% of the scores fall within two SDs of the mean.
  
• For example then, if you have a score that is one SD above the mean, then that score is at the 84th percentile, assuming a normal distribution.

Using Excel to Perform these Functions

1. Type in variable that you will be entering.

2. Enter your data in a column (for easiest visibility) underneath the variable name.

3. You are now ready to summarize your data. Before performing a function, click on the cell in which you want the result to go.

4. To perform a function, you can either go to the function key or go to 'Insert' function.

5. Choose which function you want to perform. A screen will appear that asks you to tell Excel which numbers should be used in the calculation.

6. To indicate which numbers Excel should use you have two choices. One choice is that you can type in the cell letter and number, divided by a colon. Alternatively, you can click on the icon with the small red square. After doing so, highlight the numbers you wish to be used in the calculation. Then, click on the icon with the red square again. The summary statistic should appear in the cell you indicated.

7. Using the above steps, you can calculate the mean, median, mode, minimum and maximum score (to get the range), and the standard deviation.

8. If you are interested in more advanced data analyses and they are not available on your computer, you can usually "add-in" a specific feature with the MS Office CD.

9. Randy Gabel has created a VERY HELPFUL website for first-year students, with resources supplied by our very own Ginger Montecino. The website, among other things, has in great detail how to create frequency tables and instructions for using Excel. The address for this page is http://classweb.gmu.edu/mgabel/unit1_2001/