You are on page 1of 5

Nursing Research (Week 11)

mjcy,rn,man

DATA COLLECTION METHODS Types of Data  Primary y Raw data that are collected first-hand by the investigator using tools or forms designed by him/her y Data are collected mainly to serve the purposes of the investigator  Secondary y Data that have been previously collected by another person or agency for other purposes and will be used by the investigator for his study Data Collection Instruments y Should be clearly identified and described y Scoring procedures and the range of possible scores on the instrument should also be included y It must be valid, reliable, sensitive, and specific y Pilot study results should be included for any newly developed or revised instrument y Standardized versus research-constructed y Methods of data collection y Primary ( Adv. Better quality, Dis. More time, more cost)  Observation  Extraction of Information from subjects through simple visual observation or use of apparatus  Observation of behavior or practices  Physical examination  Performing laboratory test  Requires skill, equipment or facilities  Advantages  If informants are unable to provide information (Ex. babies behavior)  If informants would give inexact answers (Ex. Blood pressure)  If people resist questions or are unwilling to answer  Can record events as they occur  Disadvantages  Impossible to predict occurrence of events  Ex. Investigator may not be around when it happens  Some event not accessible to direct observation (Ex. Sexual behavior)  Cant be used for events of the past  Cant be used in studying opinions and attitudes  Removal of Observer Bias use of camera or tape recorder  Interview  One-to-one encounter between the interviewer and respondent facilitated by the help of a list of questions which may be structured or not  Advantages:  Used with almost all sectors of population  Yield better number of respondents  Can make observations during interview  Permits clarification of questions or responses

Nursing Research (Week 11)




mjcy,rn,man

Disadvantages:  Entails greater expense  Cant reach a greater number of people  Possibility that responses can be influenced by the attitude of the interviewer  Interviewer may inject some of her opinions  Questionnaire  Contains written questions with spaces provided for the respondents to write the answers  Can be conducted in different ways:  Send questionnaire to respondents  Mail, hand deliver, channel through a point person  Gathering respondents in one place at one time  Important: Must have clear instructions on how to accomplish form  Advantages:  Can be administered to a large group simultaneously for scattered population  Confidence at anonymity  Avoids influence exerted by the interviewer  Disadvantages:  Response rates are usually low  Only for the literate  Completeness of data not always assured  Potential problem of incoherent answers y Secondary (Adv. Cheaper, less time, Dis. Less quality) y Records review  Data are taken from available records, such as:  hospital records  employment records  pathology logbooks  census  registries of vital events  reports of notifiable diseases  publications of health statistics  registries for certain diseases  data from previous studies  Advantages  Economical  No cooperation needed from subject  Can establish trends over periods of time  Maybe used as baseline data  Disadvantages  Potential problem with access  Issue of confidentiality may arise  Problem with accuracy

Nursing Research (Week 11)


   Problem with completeness Different definition of terms Data may be out-dated

mjcy,rn,man

Criteria for Selection of Data-Collection Instrument A. Objectives of the study B. Feasibility/ Practicality y Acceptability of procedure inconvenience, unpleasantness y Need for skills and training y Source y Cost y Time C. Validity/Reliability y Accurate information y Completeness

Reliability  Concerns with consistency and stability  The degree of reliability is usually determined between two sets of scores or between the ratings of two judges  Types of Reliability  Test-retest (stability) reliability  When a test is given to the same samples on two different occasions and these two tests show a high degree of correlation  Stability over time (Ex. Thermometer)  Interrater (equivalence) reliability  Comparisons of raters or observers rating using the same instrument (2 or more raters)  Internal Consistency reliability  Address the extent to which all items on an instrument measure the same variable  Split-half reliability  The most popular test of reliability  Single administration of test  Test is split into halves and scored separately  Correlation is computed Validity  The validity of instrument concerns its ability to gather the data that is intended to gather  The greater the validity of an instrument, the more confidence you can have that the instrument will obtain data that will answer the research questions or test the research hypothesis.  Types of Validity  Face Validity o An instrument is said to be valid when an examination shows that it is measuring what it is supposed to measure

Nursing Research (Week 11)


o

mjcy,rn,man

It can be examined through the use of experts in the content area or through the use of people who have characteristics similar to those of the potential subjects. o Subjective----rarely used alone Content Validity o It is concerned with the scope or range of items used to measure the variable o Methods  Comparing the content of the instrument with the available literature on the topic  Panel of experts  Developing a test blueprint designed around the objectives for the content Criterion Validity o Is concerned with the extent to which an instrument corresponds to or is correlated with some criterion measure of the variable of interest. o It assesses the ability of an instrument to determine subjects responses at the present time or predict subjects responses in the future  Does the instrument under consideration correlate with another instrument that measures the same behavior or responses? Construct Validity o It is the degree to which an instrument measures the construct it is supposed to measure  Construct is a concept or abstraction created or constructed by the researcher  The most difficult to measure o Factor analysis o Has been considered the most powerful method of construct validation Internal Validity o Is the extent to which we can be sure of the accuracy or correctness of the findings of the study o It refers to how accurate the results are within itself or internally o Six Threats to Internal Validity  Selection Bias  Changes in the dependent variable due to subject differences and not the experimental treatment  History  An event other than the experimental treatment occurs during the course of a study and influences the dependent variable  Maturation  Changes that occur within subjects during an experimental study may influence the study results  Testing  The influence of the pretest or knowledge of baseline data effects the posttest score  Instrumentation Change  Instrument inaccuracy or judging influences the pretest or posttest scores  Mortality

Nursing Research (Week 11)




mjcy,rn,man

 Attrition or drop out External Validity o Is the extent to which the results of a study can be applied to other groups or situations o It refers to how accurately the study provides knowledge that can be applied outside of or external to the study itself o Major Threats to External Validity  Hawthorne effect  Study participants respond in a certain manner, not because of the experimental treatment but because they are aware that they are being observed  Solution: Double blind experiment  Experimenter effect  Researcher characteristics or behaviors influence subject behaviors  Rosenthal Effects  Interviewer characteristics or behaviors influence respondents answers  Reactive effects of the pretest  Subjects are sensitized to the experimental treatment through taking the pretest  Placebo Effect  is a psychological response to treatment  is the phenomenon whereby a patient's symptoms can be alleviated by an otherwise ineffective treatment; most likely because the individual expects or believes that the treatment will work  is the phenomenon where people perceive symptoms or side effects from an ineffective treatment because they have been led to believe they might occur

You might also like