Vous êtes sur la page 1sur 34


Dr.Valliammal Shanmugam Lecturer CON NIMHANS

Data collection begins after a research problem has been selected and research design has been planned out.

Primary and secondary data
From where or from whom will you get the information? Existing information records, reports, program documents, logs, journals People participants, parents, volunteers, teachers Pictorial records and observations video or photos, observations of events, artwork

Types of Primary Data

Demographic/Socioeconomic Age, Sex, Income, Marital Status, Occupation Psychological/Lifestyle Activities, Interests, Personality Traits Attitudes/Opinions Preferences, Views, Feelings, Inclinations Awareness/Knowledge Facts about product, features, price, uses Intentions Planned or Anticipated Behavior Motivations Why People Buy (Needs, Wants, Wishes, Ideal-Self) Behavior Purchase, Use, Timing, Traffic Flow

Primary Data Can Be Gathered By:

Communication Methods
Interacting with respondents Asking for their opinions, attitudes, motivations, characteristics

Observation Methods
No interaction with respondents Letting them behave naturally and drawing conclusions from their actions

Communication Methods of Primary Data Collection

Methods include: Surveys Focus Groups Panels Highly versatile in terms of types of data Generally more speedy Typically more cost effective Electronic media have made observation cheaper Activities, Interests, Personality Traits


Common methods include:
Case Study Interview

Tests Photograph, videotape, slides Diaries, journals, logs Document review and analysis

Group assessment Expert or peer reviews

Portfolio reviews

Survey Research
Most common type of research in the social sciences. When to use:
Used when researcher wants to get at underlying attitudes and disposition surrounding a piece of information. Used when researcher wants to look at broad patters of social life or to discuss widespread social reactions(not necessarily the case for in-depth interview). Used to gather information to describe characteristics of a specific population.

The quality of a survey depends on its design, including questions and sample selection of population

Advantages of Survey Research

Allows researcher to select precise samples of population of interest. This results in
Generalize results
Relate finding from sample to larger population

The ability to use statistical techniques and to aggregate data so can discuss general characteristics of a social group.

The ability to test reliability of indicators (questions) The ability in panel studies to establish a time dimension to carry-out causal analysis.

Disadvantages of Surveys
Difficult to find out in-depth information about a cultural or social group. Difficult to use when unfamiliar with the population of interest. (ethnography better) Inappropriate when interested in establishing causes and processes for specific behaviors or attitudes. (experiments better)

Types of Survey Administration

Face to face Telephone Focus groups Used in other types of research including qualitative studies

Self Administered
Mail Group or in person administration Mail or internet

Types of Survey Schedules and Questions

Types of questionnaires
Structured Unstructured

Types of questions
Open-ended Closed response

Structured Interviews
Similar to self administered questionnaire.
Administer an interview schedule with specific question with mostly forced choice responses.

Different from self administered questionnaires

Requires transitional phrases for interviewer Spoken language

Cant require the respondent to remember too much information

Simplify responses Provide cards with responses Use rating scales Use card sorts for responses

Interviewers Qualities are important

Important Qualities for Interviewer

Suppress judgment and opinions Do not offer information that is not required or requested Be a good listener Establish trust and rapport Know the interview. Use standard technique for interview (ask same questions in same way). Be committed to completing every interview Understand how to probe when necessary Use common sense in difficult situations

Observation Methods: What Can Be Observed?

Physical Actions Shopping behavior, response latency, service quality, television viewing habits Verbal Behaviors Sales conversations, opinion leadership, tone of voice Expressive behaviors Facial expressions, body posture Spatial Relations and Locations Traffic patterns, store layout, efficiency Temporal Patterns Amount of time spent shopping, service time Physical Evidence Amount and type of food consumed, media read

Observation Methods of Primary Data Collection

Methods include: Direct observation Contrived observation (laboratory) Content Analysis Physiological measurement Electronic methods Greater objectivity less researcher bias More accurate less response tendency or demand effects Limited in terms of what can be observed

Focus Groups
Similar to interview in that a moderator or facilitator interview group or leads discussion. When to use
To get a focused reaction of a small group to a specific issue, item, product To get impressionistic responses about something or someone To dig deeply into one topic or area To get background information for other research method
Explore possible hypotheses Interpret findings

Most common in market research and political polling

Designing a Focus Group

Group Selection and Composition
Should be between 6-12 members. Ideal size seems to be about 8. Select individuals or recruit participants who are knowledge about subject, have experiences that are relevant or have an interest in topic. Consider the characteristics of the group members and the group dynamic of the group. Offer enticement for participation that do not sway group
Food and/or drink Sometimes prizes or awards

Designing a Focus Group

Select a moderator or group leader
Should be good facilitator Should not be judgmental Should be knowledgeable about research problem Should be able to encourage discussion Should establish equality among members and encourage diverse points of view Should be authoritative enough to keep discussion on track but not so authoritative as to stifle natural flow of conversation but not be too authoritative to as to stifle free expression of ideas

Designing a Focus Group

Qualities of Selected Environment
Non-threatening Comfortable Convenient Facilitate conversation and interaction

Types of questions
Different types of questions are used for different purposes and are appropriate at different times during a focus group (See p227 in text)

Methods are often thought of as quantitative or qualitative...

Quantitative methods Qualitative methods
Surveys Questionnaires Focus groups

Existing databases

Unstructured interviews
Unstructured observations

Quantitative and Qualitative information

"Not everything that counts can be counted."

5 (Quantity) Happy (Quality) Kids

Quantitative data collection methods produce numbers. Qualitative data collection methods produce words. Quantitative and qualitative each has its strengths and weaknesses. Quantitative methods are more structured and allow for aggregation and generalization. Qualitative methods are more open and provide for depth and richness.

What method shall I use?

There is no simple answer There is no ONE best method It all depends

When choosing methods, consider

The purpose of your evaluation Will the method allow you to gather information that can be analyzed and presented in a way that will be credible and useful to you and others?

The respondents What is the most appropriate method, considering how the respondents can best be reached, how they might best respond, literacy, cultural considerations, etc.?

What kind of data your professionals will find most credible and useful Resources available. Time, money, and staff to design, implement, and analyze the information. What can you afford? Type of information you need. Numbers, percents, comparisons, stories, examples, etc.

Interruptions to program or participants. Which method is likely to be least intrusive? Advantages and disadvantages of each method. .The importance of ensuring cultural appropriateness.

Often, it is better to use more than one data collection method.

Why would this be so?

When we use several methods we say we are triangulating. Triangulation is important in evaluation because we want accurate and trustworthy information.
Triangulation means the use of multiple sources and methods to gain a better understanding. Each source and each method has inherent biases so using more than one source and/or method provides a more accurate picture.

How might you mix sources of information in your evaluation? How might you mix data collection methods to evaluate your program?

Instrument to collect the information

What will you use to actually collect the data a recording sheet, a questionnaire, a video or audio tape?

The term, instrument , sounds like we are talking about a dental office, a cockpit or an orchestra.
Actually, we use the term instrument to mean the tool on which the data is actually recorded: the questionnaire, the recording form, the video or audio tape, for example. If you have selected a survey as your method, you automatically know that you will need a questionnaire. But, if you choose a method such as focus group or interview or observation, think about what you will use for recording the information.

Choices: Timing
When Will Data be Collected? Before and after the program At one time At various times during the course of the program Continuously through the program Over time - longitudinally

Reflection time
What is one thing you learned (or had reinforced) from going through this presentation that you hope not to forget? Good luck with your data collection efforts!