Methods Of Data Collection

MET HO DS OF DATA COLLE CT IO N

METHODS OF DATA COLLECTION Like research design, there are many

alternative approaches to data collection, and these approaches vary along several dimensions.

EXISTING DATA VERSUS ORIGINAL DATA One of the first decisions that

investigators make with regards to research data concerns is whether to use existing data or to collect data generated specifically for the study.

HISTORICAL RESEARCH typically relies exclusively on available

data is the systematic collection of data relating to past occurrences data for historical research are usually in the form of written resources of the past: periodicals, diaries, letters, newspapers, minutes of meetings, legal document, reports and so forth

HISTORICAL RESEARCH The historical researcher usually must

evaluate the authenticity and accuracy of historical data before analyzing them.  Example: Widerquist(1992) studied Florence Nightingale’s spirituality and its influence on the development of modern nursing through an analysis of Nightingale’s letters, diaries, essays and journals.

SECONDARY ANALYSIS  – the use of data gathered in the previous study to test

new hypotheses or address new research questions.  The difference between using records and doing a secondary analysis is that the researcher performing a secondary analysis typically has an intact data set that is ready to analyze, using records one has to assemble the data set from records and considerable coding and data manipulation are usually necessary.  The primary advantages of using existing data are that they are economical and time saving.  The collection of original data is typically costly and time concerning.

TH REE TYPES O F APPRO ACH to DATA CO LLECTI ON Self Report Observation Biophysical

Regardless of what specific approach is used data collection methods vary along several important dimensions.

KEY DI MENS IONS OF DATA COLLE CTION ME THODS developing a plan for data collection, the investigator makes many important decisions.

KEY DIMENSIONS OF DATA COLLECTION METHODS: STRUCTURE include a fixed set of questions that are

generally answered in a specified sequence and with pre designated response options Example: agree or disagree in structured methods, there is little opportunity for participants to qualify their answers or to explain the underlying meaning of their responses

UNSTRUCTURED  yield data that are considerably more difficult to

analyze. Example: Structured: During the past week, would you say you felt stressed?  rarely or non of the time  some or a little of the time  occasionally or a moderate amount of time  most or all of the time Example: Unstructured How stressed or anxious have you been this past week? Tell me about the kind of tension and stresses you have been experiencing.

QUANTIFIABILITY data that will be subjected to statistical

analysis must be gathered in such a way that they can be quantified for statistical analysis, all variables must be quantitatively measured - even though the variables are abstract and intangible phenomena that represents qualities of human, such as hope, loneliness, pain and body image

OBTRUSIVENESS data collection methods differ in terms

of the degree to which people are aware of their status as study participants If participants are fully aware of their role in a study, their behavior and responses may not be normal.

OBJECTIVITY refers to the degree to which two

independent researchers can arrive at similar “scores” or make similar observations regarding the concepts of interest, that is, make judgment regarding participants attitudes or behavior that are not biased by personal feelings or beliefs

MAJOR TYPES OF DATA COLLE CTION MET HOD S

SELF REPORT METHODS A good deal of information can be

gathered by questioning people. Self Report data can be gathered either by oral interview or by written questionnaire.

UNSTRUCTURED AND SEMISTRUCTURED SELF-REPORT TECHNIQUES

Unstructured or loosely structured self

report methods offer the researcher flexibility in gathering information from study participants. Unstructured or semi-structured interviews in other words tend to be conversational and interactive in nature

TYPES OF UNSTRUCTURED SELF REPORTS 1. Completely unstructured interview  are used when the researcher proceeds with no preconceived view of the specific content or flow of the information to be gathered  the aim of these interviews is to elucidate the respondents perception of the world without imposing on them any of the researchers views  A researcher using a completely unstructured approach may internally ask a broad question (sometimes called a grand tour questions)  Example: Tell me about what happened when you first learned you had AIDS?

2. Focused interview or semi-structured interview  the interviewer uses a list of areas or questions to be covered with each respondent  the list is referred to as a topic guide  most widely used method of collecting unstructured self report data

3. Focus group interview  a group usually 5 to 15 people is assembled for a group discussion  the interviewer (often called a moderator) guides the discussion according to a written set of question or topic to be covered.

4. Life histories  are narrative self-disclosures about a persons life experiences  this approach, the researchers ask the respondents to provide on chronologic sequence, a narration of their ideas and experiences regarding some terms, either orally or in writing  Example: some researchers have used this approach to obtain a total life health history

5. Critical incident technique  is a method of gathering information about people’s behaviors by examining specific incidents relating to the behavior under investigation  the word critical means that the incident must have a discernible impact on some outcome, it must make either a positive or negative contribution to the accomplishment of some activity of interest  the technique differs from other self report approaches is that it focuses on something specific about which the respondent can be expected to testify as an expert witness

6. Diaries  ask participants to maintain a daily log concerning some aspect of their lives over a specified period of time

GATHERING UNSTRUCTURED SELF REPORT DATA  The purpose of gathering unstructured self report

data is to enable the researcher to construct reality in ways that are consistent with the construction of the people being studied.  This goal requires the researcher to take steps to overcome communication barriers and to enhance the flow of meaning.  An important issue is that the researcher and the respondents should have a common vocabulary

In addition to being good questioners, the

researchers must be good listeners. Only by attending carefully to what the respondent is saying can the in-depth interviewer develop appropriate follow-up questions. Even when a topic guide is used, the interviewer must not let the flow of dialogue be bound by those questions; many questions that appear on a topic guide are answered spontaneously over the course of the interview, usually out of sequence.

GATHERING UNSTRUCTURED SELF REPORT DATA  Unstructured interviews are typically quite long –

sometimes lasting up to several hours. Researchers often find that the respondents’ construction of their experience only begins to emerge after lengthy, in-depth dialogues. The issue of how best to record such abundant information is difficult one. Some researchers take sketchy motes as the interview progresses, filling in the details as soon as practical after the interview is completed. Many prefer tape recording the interviews for later transcription.

STRUCTURED SELF REPORT INSTRUMENTS  A researcher using a structured approach

always operates with a formal written instrument known as an interview schedule.  The instrument or sometimes an SAQ (Selfadministered questionnaire)  Structured instruments consist of a set of questions (also known as items) in which the wording of both the question and in most cases, the response alternatives is predetermined

QUESTION FORM Structured instruments themselves vary in

their degree of structure through their combination of open-ended and closedended questions.

Open-ended questions allow respondents to respond in their

own words open-ended questions give a lot of freedom to the respondent and therefore, offer the possibility of spontaneity which is attainable when a set of responses is provided Example: What was the biggest problem you focused after your open

Closed-ended or fixedalternative questions Offer respondents a number of

alternative replies from which the subjects must choose the one that most closely matches the appropriate answer. The alternatives may range from the simple yes or no variety

SPECIFIC TYPES OF CLOSED ENDED QUESTIONS 1. Dischotomous questions – require the respondent to make a choice between two response alternatives such as a yes or no or male or female 2. Multiple choice – after more than two response alternatives - Multiple choice questions most commonly offer three to seven alternatives 3. Cafeteria questions – are a special type of multiple choice questions that asks respondents to select a response that most closely corresponds to their view.

4. Rank order questions – ask respondents to rank target concepts along some continuum, such as most favorable to least favorable or most to least important - Rank order questions can be useful but need to be handled carefully because they are often misunderstood by the respondents - Rank order questions should not ask respondents to rank more than about 10 alternatives 5. Forced-choice question – requires respondents to choose between two alternative statements that represent polar positions or characteristics 6. Rating questions – ask respondents to judge something along an ordered dimension - require labeling of the end points but sometimes intermediary points along the scale are also labeled

7. Checklist – is often a two-dimensional arrangement in which a series of question is listed along as one dimension. 8. Calendar questions – used when researcher want to obtain retrospective information about the chronology of the different events and activities in peoples lives 9. Visual Analogue – is a straight line, the end anchors of which are labeled as extreme limits of the sensation as feeling being measured - Subjects are asked to mark a point on the line corresponding to the amount of sensation experienced.

INTERVIEWS VERSUS QUESTIONS An important decision that the researcher

must make when using a structured selfreport approach concerns the use of an interview versus a questionnaire.

ADVANTAGES OF QUESTIONNAIRES:  Questionnaires are much less costly and

requires less time and energy to administer.  Questionnaires offer the possibility of complete anonymity, which may be crucial in obtaining information about illegal, immoral, or deviant behaviors or about embarrassing characteristics.  The absence of an interviewer ensures that there will be no bias in the responses that reflect the respondent’s reaction to the interviewer rather than to the question themselves.

ADVANTAGES OF INTERVIEW:  The response rate tends to be high in face-to-face

interviews. Respondents are generally more reluctant to refuse to talk to an interviewer than to ignore a questionnaire, especially a mailed questionnaire. Low response rates can lead to serious biases, because people who complete the questionnaire or interview are rarely a random subset of those whom the researcher intended for inclusion in the study.  Many people simply cannot fill out a questionnaire; examples include young children, the blind, and the very elderly. Interviews are feasible with most people.

Interviews are less prone to

misinterpretation by the respondents because the interviewer is present to determine whether questions have been misunderstood. Interviewers can produce additional information through observation. The interviewer is in a position to observe or judge the respondent’s level of understanding, degree of cooperativeness, lifestyle, and so on. These kinds of information can be useful in interpreting responses

Scales – a scale is a device designed to assign

a numeric score to people to place them on a continuum with respect to attributes being measured.

Likert scales  most widely used technique  named after the psychologist Rensis Likert  it consist of several declarative items that express a

view point on a topic  respondents are asked to indicate the degree to which they agree or disagree with the opinion expressed by the statement  The researcher would, therefore, assign a higher score to a person agreeing with this statement than to someone disagreeing with it.  Because the item has five response alternatives, a score of 5 would be given to someone strongly agreeing, 4 to someone agreeing, and so forth.

Semantic Differential (SD)  respondents is asked to rate a given concept  example: primary nursing, team nursing, on a

series of bipolar adjectives such as good or bad, strong or weak, effective or ineffective, important or unimportant.  the scoring procedure for SD responses is essentially the same as of Likert scales. Scores from 1 to 7 are assigned to each bipolar scale response with higher scores generally associated with the positively words adjective.  Scales permit researchers to efficiently quantify subtle gradations in the strength or intensity of individual characteristics.

CRITIQUING SELFREPORTS One of the first questions a consumer must

ask about the data collection method of a self-report study is whether the researcher made the correct decision in obtaining the data by means of self-report rather than by an alternative method.

It may be difficult to perform a thorough critique of self report methods in studies that are reported in journals because a detailed description of the data collection methods may not be included. What the reader can expect is information about the following aspects of the self-report data collection:  

 





The degree of structure used in the questioning Whether interviews or questionnaires (or variants such as a projective method or Q-sort) were used Whether a composite scale was administered The length of time it took, on average, to collect data from each participant How the instruments were administered (e.g., by telephone, in person, by mail, and so forth) The response rate

 The degree of structure that the researcher

imposes on the questioning is of special importance in assessing a data collection plan. For example, respondents who are not very articulate are more receptive to structured instruments with many closed-ended questions than to questioning that forces them to compose lengthy answers. Other considerations include the amount of time available; the expected size of the sample (open ended questions and unstructured interviews are difficult to analyze with large samples); the status of existing information on the topic (in a new area of inquiry, a structured approach may not be warranted); and, most important, the nature of the research question.

REFERENCES: Polit, D. & Hungler, B. (1999). Nursing

research: Principles & methods. Lippincott, Philadelphia. Polit, D. & Hungler, B. Essentials of nursing research methods, approach, and utilization. 4th ed. Lippincott Polit, D. & Beck, C.T.(2004). Nursing research 7th ed.Philadelphia:Lippincott

THANK YOU!!! HAVE A BLESSED CHRISTMAS AND A PROSPEROUS NEW YEAR!!!