Day 7

Table of Contents

Review

Qualitative research is "the collection and analysis of extensive narrative date to gain insights into a situation of interest not possible using other types of data." Gay, p 208.

Historical Research:

Describes causes, effects, or trends that may help explain present events and anticipate future events

Qualitative Research:

Attempts to ascertain the "why" by examining the feelings, beliefs, and meanings people attach to the object under study

Qualitative Instrument:

The research acts as the instrument to gather narrative data in naturalistic setting. Observations, interviews, and document collection are common.

Biases:

Biases come from the beliefs of the observer, and the altered actions of those being observed.

Analysis:

Look for categories or themes. Include direct quotes to capture feelings, attitudes, and beliefs. Conclusions are tentative and can't be generalized.

Definition and purpose of descriptive research

Descriptive research involves collecting numerical data to test hypotheses or answer questions concerning current status.

Conducted either through self-reports collected through questionnaires or interviews (person or phone), or through observations.

Example: A study to determine the beliefs of EdTec graduate students.

Common characteristics:

Often requires the creation of an instrument since the researcher is looking for new knowledge.
Sample is usually random or stratified-random to be representational and generalizable, or census.

Subtypes of common descriptive studies:

Cross-Sectional: taken at one time--a snapshot of the participants
Longitudinal: sample followed over time to see how they change
- Trend: annual selection of new people within a certain group
- Cohort: Different participants each year, but from the same initial group
- Panel: Same people each year, from within the same initial group
- Follow-up: The panel group studied some time later

Self-reported descriptive research

Self-reported studies require the collection of standardized, quantifiable information from members of sample or population.

Questionnaires are most common method. Questionnaires should:

Be attractive, brief, and easy to respond to
Use predominately close ended questions
Not include leading questions
Not include questions that assume a fact not in evidence
Avoid jargon
Be specific
Each question on a specific concept
Be validated before their use
Typical types are scaled items, ranked items, checklist items, and free response items.

Introduce your questionnaire with a cover letter:

Explain purpose and significance to motivate participants
Be brief, neat, and addressed specifically to the person
Include an endorsement of the organization or administrator
Assure confidentiality
Include specific return date, and stamped, addressed envelope.

Follow-up as soon as possible with non-respondents.

Postcard soon after closing date
Second questionnaire with new letter and envelope

Questionnaire Activity

Let's suppose you are working on a project to identify how often, and for what purposes, EdTec graduate students are using their home computers.

Your task is to create valid questions for a questionnaire. What would those questions include?

You have 15 minutes to generate questions (and be ready to justify each question) that consider the previous points as to what a good questionnaire includes.

Interviews

Interviews are another type of self-reported studies.

Advantages over questionnaires is that they produce more in-depth data due to followup questions.

Disadvantages include that questionnaires:

Need a written guide for interviewers
- Must follow same procedures each time (such as what prompting is allowed)
Are slow and expensive; therefore smaller sample size will be used

Tips for good interview results:

Begin by building rapport with participant
Use a recording device
Pretest with sample population
Start with structured questions, then move to unstructured

Phone calls can be used for interviews for quicker data collection and expanded sample

Observational research

Observations are the second major division of collecting data for descriptive statistics, normally done in a naturalistic setting, or through role playing in a simulation

Sequence of events in setting up an observation include:

Define what specific behaviors match the intended behavior (how would you operationalize constructivism?)
Establish code or set of symbols and recording device (checklist)
Train observers on what to observe, how to record, and how often to record. Check for reliability (80% agreement)
Specify observational period selecting different days or times
Best to observe and record only one behavior at a time

Biases may appear due to:

An observer rating a majority of observees as above, average, or below regardless of actual behavior
The initial impressions concerning an observee affecting subsequent observations (halo effect)
Observee reacts atypically because they are being observed (observer effect)

Descriptive statistics

Measures of central tendency and measures of variability are two types of descriptive statistics. Each type of measures has an advantage in providing information to the reader.

Measures of Central Tendency (identify the "middle")

The Mean: I mean the average
The Mode: The most
The Median: The middle item

Measures of Variability (measure the dispersion)

The Range: The difference between highest and lowest score
The Quartile: Divide the factors into four groups to identify their location
Variance: The amount of spread among test scores. Each score, minus the mean, squared, totaled, and divided by number of scores.
The Standard Deviation: How close the numbers gather around the mean. The square root of the variance.

Closure; Review and Assignments

Review questions: (To find the answers, click the question mark icon)

What is, and what is the purpose of, descriptive research?
In self-reported research, what are the advantages and disadvantages between questionnaires and interviews?

What are the major constraints in doing observational research?

Explain the various statistical methods to identify central tendency and variability

Before next week:

Read Chapter 9 and pages 443-451 in book

Historical Research:	Describes causes, effects, or trends that may help explain present events and anticipate future events
Qualitative Research:	Attempts to ascertain the "why" by examining the feelings, beliefs, and meanings people attach to the object under study
Qualitative Instrument:	The research acts as the instrument to gather narrative data in naturalistic setting. Observations, interviews, and document collection are common.
Biases:	Biases come from the beliefs of the observer, and the altered actions of those being observed.
Analysis:	Look for categories or themes. Include direct quotes to capture feelings, attitudes, and beliefs. Conclusions are tentative and can't be generalized.