RESEARCH

Research guides nursing practice to improve care and quality of nursing intervention

Paradigm – world view, disciplined inquiry in the field of nursing2 Types:1. Positivist Paradigm

- assumes that there is an objective reality and that natural phenomena are regular and orderly

2. Naturalistic Paradigm- assumes that reality is not a fixed entity but rather a construction of human mind.

General Types:1. Descriptive Research

- answers questions who, what, where, when.- Its purpose is to observe, describe and document aspects of observation

Examples: “Tardiness and absenteeism among high school students” “The medicinal components of five kinds of Philippine backyard plants” “Smoking habits of health service providers in government and private hospitals”

2. Explanatory -explores the interrelationships among variables of interest without any intervention on the part of

the researcher.Examples:

“Knowledge about Cancer and Compliance with Diet, Exercise and Medical Regimen among Cancer Patients”

“Relationship Between Socioeconomic Factors and absenteeism among High School Students in District X”

2.1. Descriptive Correlational Study- Researcher is interested in describing relationships among variables, without

necessarily seeking to establish a causal connection.

3. Experimental Research- Investigator controls (manipulates) the independent variable and randomly assigns subjects on different conditions.

Examples: “The Effect of Verbal Suggestion on Overt Pain Reaction of Selected Post-Operative

Patients” “The Effect of Different Levels of Applied Nitrogen on the Growth and Yield of Rice”

Other Classifications:1. Pure or Basic Research

- to accumulate information, extending base of knowledge- to improve understanding, formulate or refine a theory

Examples: “ Attitudes Towards Health and Smoking Habits of Health Service Providers”

2. Applied Research- finding an immediate solution to an existing problem.

Examples: “The Effect of Gender Sensitivity Training on Men’s Involvement in Child Care” “ Remedial Teaching: It’s Effect on the Performance of Slow Learners”

* Basic research is appropriate for discovering general principles of human behavior and biophysiologic process, but applied research is designed to indicate how these principles can be used to solve problems in nursing practice.

1

3. Exploratory Research- investigates the full nature of the phenomenon and the other factors with which it is related.

Examples:“Menopause: Working Women’s Perceptions, Experiences and Coping Strategies”“Domestic Violence: Ideas, Experiences, and Needs of Married Working Men in the City of

Pampanga”

4. Quantitative Research- The study of the phenomena that lend themselves to precise measurement and quantification

Example:“Health Seeking Behavior and Health Status of Retired School Teachers in Zamboanga”

5. Qualitative Research- the investigation of phenomena, typically in an in-depth and holistic fashion, through the collection of rich narrative materials using a flexible research design.

Example: “Underground Economy: A Survival Strategy of Public School Employees” “ Menopause: Women’s Perceptions and Experiences”

RESEARCH METHODS:1. Experimental Method2. Survey Method3. Historical Method4. Content Analysis

THE RESEARCH PROCESS1. Identify & Define a research problem2. State a research problem / objectives3. Theoretical / Conceptual Framework4. Operational Definition of Variables5. Hypothesis Formulation6. Choosing of Appropriate Research Design7. Identification of Target Population & Sampling8. Data Collection

- Preparation of Research Instrument- Reliability Testing and Validation- Questionnaire administration, Interview, Testing and Observation- Quality Control

9. Data Processing- Editing, Coding, Encoding, Creation of Data Files, Tabulation

10. Data Analysis and InterpretationStatistical Analysis, Interpretation, Generalization

11. Report Preparation, Information Dissemination12. Recommendation

A. Problem Identification

Example: “Does the students’ use of the internet affect their performance in school?”

Characteristics of a Good Research Problem1. A good problem is relevant2. A good problem is clear3. A good problem is feasible4. A good problem is ethical

4.1. Principle of Beneficence4.1.1. Freedom from Harm (physical, emotional, psychological)4.1.2. Freedom from Exploitation

- participants need to be assured that their participation will not be used against them.

2

4.1.3. Risk/Benefit Ratio- should focus on significant topic that has the potential to improve client care

4.2. Principle of Respect for Human Dignity4.2.1. Right to Self-Determination

- right to decide voluntarily, whether to participate in a study.- freedom from coercion (explicit or implicit threats of penalty)

4.2.2. Right to Full Disclosure- the researcher has fully described the nature of the study, the person’s right to refuse participation, the researcher’s responsibilities, and the likely risks and benefits that would be incurred.

4.3. Principle of Justice4.3.1. Right to Fair Treatment

- nondiscriminatory selection, based on research requirements and not on vulnerability- nonpredjudicial treatment for those who withdraw from participating after agreeing to participate.

4.3.2. Right to Privacy- Anonymity (unlinking a participant with his/her data)- Confidentiality (info will not be publicly reported or made accessible to parties not involved in research

6. The researcher should be qualified and interested to do the study

Vulnerable Groups- those incapable of giving informed consent

1. Children – legally and ethically, obtained from parents and guardians.2. Mentally or Emotionally disabled people – cannot weigh risks and benefits in participation3. Physically disabled people – special procedures may do4. Institutionalized people – emphasize voluntary nature of participation5. Pregnant women – safeguard the mother and fetus, unless the purpose is to meet the health

needs of the woman

Related Literature - ascertain what is already known in relation to a problem of interest.♦ Primary Source – research reports which are descriptions of studies written by researchers who conducted them♦ Secondary Source – descriptions of studies prepared by someone other than the original researcher

B. Formulation of Hypothesis

Research Objectives / Aims- statements of what the researcher intends to do

Example: “To determine the extent of High school students’ participation in school activities” “To test the effectiveness of Oresol in the treatment of diarrhea”

Types of Objectives:1. General Objectives

- states clearly what the researcher will do and expects to find out

Example:Research Title: ”A Study on the Extent of Participation in School Activities of High School

Students in City A”Objective: “A survey will be conducted to determine the extent of participation in school

activities of high school students in City A during School Year 2001-2002.

What will be done? A Survey will be conductedFor what purpose? To determine the extent of participation in school activitiesWho will be studied? High School StudentsWhere? City A

3

When? School year 2001-2002

2. Specific Objectives- viewed as sub-objectives or small particles- expressed in measurable terms- suggest the type of analysis to be done

HYPOTHESIS - a tentative prediction about the relationship between two or more variables in the

variables under study.

Types:1. Null Hypothesis (Statistical Hypothesis)

- states NO relationship between Independent and dependent variableExample:

“There is no significant relationship between mass media exposure and attitude towards land reform among lowland farmers”

“There is no significant difference between the mean age of male faculty members and the mean age of female faculty members”

2. Alternative Hypothesis (a.k.a. Research, Substantive, Declarative, Scientific)- statements of expected relationships between variables.

Example: “There is a significant relationship between mass media exposure and attitude towards

land reform among lowland farmers

3. Directional Hypothesis - specifies the expected direction of the relationship between variables- usually derived from theories

*A positive or direct relationship is present when one variable increases with the increase in the value of another.

*A relationship is negative or Indirect when the value of one variable increases as the other value decreases

Example: “The higher the level of exposure of farmers to mass media the more favorable their attitude towards land reform” (positive)

“The more time employees spend in meetings, the less productive they are” (negative)

4. Non-directional Hypothesis - does not stipulate the direction of relationship (whether positive or negative)

4

Example:“ Relationship Between TV Viewing and Academic Performance of Grade Six Pupils in

Private and Public Elementary School in Region VI”

General Objective:The study will be conducted to determine the existence and degree of the relationship

between TV Viewing and Academic Performance of Grade Six Pupils in Private and Public Elementary School in Region VI

Specific Objectives:Specifically, the study aims to:

1. determine whether there is a significant relationship the pupil’s frequency of viewing TV and their general average in all subjects in grade six.

2. determine whether there is significant relationship between the amount of time spent by the pupils in viewing TV and their general average in all subjects in grade six.

Examples: “There is a difference in the level of anxiety of pre-surgical patients who receive pro-operative instruction than those who do not receive such instruction”

*Hypothesis are never proved through they are accepted or supported or rejected.

Theory – systematic, abstract explanation of some aspect of reality. *Lazarus and Folkman’s Theory of Stress and Coping

- explains people’s methods of dealing with stress, posits that coping strategies are learned, deliberate responses to stressors that are used to adapt to or change the stressors.

*Azjen and Fishbein’s Theory of Reasoned Action- provides framework for understanding the relationships among a person’s

attitudes, intentions, and behaviors. Behavioral intentions are the best predictor of a person’s behavior, and behavioral intentions are a function of attitude towards performing the behavior and subjective norms.

Model – symbolic representation of concepts and variables, and interrelationships among them.* Becker’s Health Belief Model

- health-related behavior is influenced by a person’s perception of a threat posed by a health problem as well as the value associated with actions aimed at reducing the threat

Framework – conceptual underpinnings of the study

Variable – something that varies (changes)Examples of Variables:

- Sex - revenue- Age - height- Income - weight

Constant – fixed, does not changeEx: Everyone in the group has a height of 5’4”

Continuous Variables – presented on continuum - Ex: Height, Weight, Age

Categorical Variables – placed on categories - civil status, gender

Dichotomous Variable – only 2: Ex. SexPolychotomous Variable – many answers: Ex: Civil StatusExtraneous Variable – confounding, uncontrolled variable

Types of Variables:1. Independent Variable - the PRESSUMED CAUSE of the problem.

2. Dependent Variable – the PRESSUMED EFFECT of the problem.

Example: “The Relationship Between Exposure to Mass Media and Smoking Habits among Young Adults”

IV – Exposure to Mass MediaDV – Smoking Habits

3. Intervening Variable (mediating variable)- mediates or acts like a “go-between” in a chain linking two other variables

Example: “Knowledge of the Dangers of Smoking, Attitudes towards Life, and Smoking Habits of Young Professionals”

IV – Knowledge of the Dangers of SmokingMV – Attitudes towards LifeDV – Smoking Habits

5

4. Antecedent Variable Example: “Extent of Exposure to Print Media and Reading Ability of College Freshmen”

RESEARCH DESIGN – overall plan for addressing a research question, including specifications for enhancing the integrity of the study

Qualitative Research Design- makes ongoing decisions reflecting what has already been learned

Characteristics:- flexible and elastic- merges various data collection strategies- requires researcher to be involved, and as a research instrument- calls for the researcher to be a bricolage( adept at performing number of diverse

tasks)- usually non-experimental, rarely controls or manipulates Independent variable.

3 Phases:1. Orientation and Overview – researchers enters “not knowing what is not known”2. Focused exploration – scrutiny and in-depth exploration3. Confirmation and closure – establishing trustworthy findings

Qualitative Research Traditions:Ethnography – provides framework for studying the meanings, patterns, and experiences of a

defined cultural group in a holistic fashion, aims to learn from members of the culture.

*Emic perspective – the way members of the culture envision the world, insiders view *Etic perspective – outsiders’ interpretation of the experiences of that culture.Ethnoscience – or known as Cognitive Anthropology, emphasis on semantic rules and shared

meaningsPhenomenology – concerned with the lived experiences of humans.Hermeneutics – uses lived experiences of people as a tool for better understanding the social,

cultural, political, and historical context in which those experiences occur.

Quantitative Research Design

Classification:A. Experimental Design

- researcher is an active agent rather than passive observer- tests hypothesis of cause-and-effect relationships

3 Properties:1. Manipulation – experimenter does something to participants in the study2. Control – experimenter introduces control over the experimental situation. * control group is used as a basis for evaluating the performance of experimental group.3. Randomization – participants are given equal chance to be selected.

Types of Experimental Design:1. Post-test Only Design / After Only Experimental Technique

- random assignments to 2 groups, collection of data after intervention

2. Before and After Experimental Technique / Pretest-Posttest- baseline measure of dependent variable- outcome measure of dependent variable

*involves observation of dependent variable at two points in time.

6

Age, Sex, Civil Status Exposure to Print Media

Reading Ability

Antecedent Variable Dependent VariableIndependent Variable

3. Factorial Design – manipulates two or more variablesEx: Comparing effects of tactile stimulation versus auditory stimulation on cardiac responsiveness of premature infants.

4. Crossover Design (repeated-measures design) – exposure of the same study participants to more than one treatment - possible carryover effects

Hawthorne effect – effect on dependent variable resulting from subjects’ awareness that they are participants under study.

Double Blind Experiments – neither subjects nor those administering the treatment know who is the experimental or control group.

B. Quasi-Experimental Designs - involves manipulation of independent variable, but lacks randomization or control group features

Types:1. Nonequivalent Control Group Design

- involves treatment and two or more groups of subjects observed before and after its implementation.

Example: We are to study the effect of primary nursing on staff morale in a large metropolitan hospital. Because the new system of nursing care delivery is being implemented throughout the hospital, randomization is not possible. Therefore, we decide to collect comparison data from nurses in another similar hospital that is not instituting primary hospital. We decide to gather data on staff morale in both hospitals before implementing primary nursing delivery system (pretest) and again after its implementation in the first hospital (posttest).

*Without randomization, it cannot be assumed that experimental and comparison groups are equivalent at outset.

2. Time-Series Design – control group is present, but without randomization.- collection of data over an extended time period and the introduction of the treatment during that period.

- strengthens a researchers ability to attribute change to the intervention.

C. Non-Experimental Design - uses ex post facto or correlational research; basically, to study relationships among variables.

-historical approach – relies on available data; data are in the form of written, narrative records of the past: diaries, letters, newspapers, minutes of meeting, reports, etc.

-descriptive approach

Retrospective Studies – investigator focuses on a presently occurring outcome, and then ascertain the antecedent factors that have cause it.

Ex: In Lung Cancer Research, the investigator begins with a sample of those who have lung cancer and those who do not. The researcher looks for differences in antecedent behaviors or conditions, such as smoking habits.

Prospective Studies –starts with presumed cause and then go forward to the presumed effect.

Ex: With the above example, researcher may begin with samples of smokers and non-smokers and later compare 2 groups for lung cancer incidence.

Cross-sectional Design – collection of data at one point in time.

7

- appropriate for describing the phenomena - economical and easy to manage

Ex: A researcher might study whether psychological symptoms in menopausal women are correlated contemporaneously with physiologic symptoms.

Longitudinal Designs – designed to collect data over an extended time period - multiple data collection points

a. Trend Studies – samples from gen. population are studied over timeb. Panel Studies – same participants supply the data at two or more points in timec. Follow-up Studies – determines the subsequent status of subjects with a specified

condition or those who have received specified intervention

Example: Patients who received a particular nursing intervention may be followed up to assess the long term effects of treatment.

SPECIFIC TYPES OF QUANTITATIVE RESEARCHa. Surveys – obtaining information about variables of a given population at a specific time.

- superficial, suited for extensive rather than intensive analysis3 Most Common Methods1. Personal, Face-Face interviews2. Telephone Interview3. Self-administered questionnaires

b. Evaluations – used to find out how well a program, treatment, practice or policy works.

Types of Evaluation:1. Process Analysis (Implementation Analysis)

- obtain descriptive info about the new program2. Outcome Analysis

- documents the extent to which goals of a program occur (positive outcomes occur)3. Impact Analysis

- identifies the impact of an intervention

Outcomes Research - documents the effectiveness of health care services.

Methods of Controlling Subjects Characteristics:1. Randomization

- to secure comparable groups, to equalize groups with respect to extraneous variables

2. Homogeneity - only subjects who are homogenous with respect to the extraneous variables are

included in the study

3. Matching - using info about subject characteristics to form comparison groups.

4. Statistical Control

SAMPLING – process of selecting a portion of the population to represent the entire population

Representativeness – the extent to which the sample is similar to the population and avoids bias.

Sampling Bias – systematic overrepresentation or underrepresentation of some segment of the population.

Types of Sampling:

1. Non-Probability Sampling - elements are selected non-randomly

8

Random assignment means that subjects are given the equal chance to be selected.

1.1 Accidental Sampling (Convenience Sampling)- uses available people as study participants- weakest sampling form

Ex: A nurse distributes questionnaires about breastfeeding intentions to the first 100 available pregnant women.

1.2. Snowball Sampling (or Network Sampling / Chain Sampling)- early sample members are asked to refer others who meet the study’s eligibility criteria

Ex:A researcher is studying environmental engineers but can only find five. She asks these engineers if they know any more. They give her several further referrals, who in turn provide additional contacts. In this way, she manages to contact sufficient engineers.

1.3. Quota Sampling- uses knowledge about the population to build some representativeness into the sampling plan - researcher identifies strata of the population and specifies the number of elements needed for those strata.- you select people nonrandomly according to some fixed quota

ExampleA researcher in the high street wants 100 opinions about a new style of cheese. She sets up a stall and canvasses passers-by until she has got 100 people to taste the cheese and complete the questionnaire.

2 Types:1.Proportional Quota Sampling you want to represent the major characteristics of the population by sampling a proportional amount of each. For instance, if you know the population has 40% women and 60% men, and that you want a total sample size of 100, you will continue sampling until you get those percentages and then you will stop. So, if you've already got the 40 women for your sample, but not the sixty men, you will continue to sample men but even if legitimate women respondents come along, you will not sample them because you have already "met your quota." The problem here (as in much purposive sampling) is that you have to decide the specific characteristics on which you will base the quota. Will it be by gender, age, education race, religion, etc.?

ExampleIt is known that 90% of nurses in a region are women. A study with a sample size of 200 nurses thus selects 180 female nurses and 20 male nurses.

2. Nonproportional quota sampling is a bit less restrictive. In this method, you specify the minimum number of sampled units you want in each category. here, you're not concerned with having numbers that match the proportions in the population. Instead, you simply want to have enough to assure that you will be able to talk about even small groups in the population. This method is the nonprobabilistic analogue of stratified random sampling in that it is typically used to assure that smaller groups are adequately represented in your sample.

ExampleA study of the prosperity of ethnic groups across a city, specifies that a minimum of 50 people in ten named groups must be included in the study. The distribution of incomes across each ethnic group is then compared against one another.

1.4. Purposive Sampling (Judgemental Sampling)

9

- based on the knowledge that a researcher’s knowledge can be used to handpick the cases to be included in the sample.- researchers might decide to purposefully select subjects who are judged typical or particularly knowledgeable of the subject matter.

MethodWhen taking the sample, reject people who do not fit a particular profile.

ExampleA researcher wants to get opinions from non-working mothers. They go around an area knocking on doors during the day when children are likely to be at school. They ask to speak to the 'woman of the house. Their first questions are then about whether there are children and whether the woman has a day job.

Other Types of Sampling:1. Modal Instance Sampling – focus on ‘typical’ people

UseUse when you want to investigate thoughts and actions of 'typical' people and when you fear that significant data about this group of people might be lost in a more general study.

MethodIdentify what 'typical' means. This could be done by a preliminary study to find the best variables to use and the most common values for these.Select people for further study based on these criteria.

ExampleA researcher wants to study the typical video game user. They do an initial study of a wide range of people and find that the majority of people playing video games are prosperous males between the ages of 18 and 25. They then recruit only people who fit this criteria to do their study.

DiscussionThe mode is the most common item in a population. For example the most common international male shoe size is 43. The mode can thus be considered as typical of a population.In practice, what is modal (or 'typical') is often based more on the views of the researcher than an extensive pre-study. For example, a TV reporter may seek 'typical weekday shoppers' and hence pick on conventionally-dressed women who look between the ages of 25 and 35 and who are pushing a pram (note how these can be judged from appearance only, making the job easier).

2. Expert Sampling – selecting ‘experts’ for opinion or study

UseUse when the research requires assessment or opinions of people with a relatively high level of skill or knowledge.

MethodIdentify what 'expert' means, for example using specific qualifications and experience. This could be done by a separate pre-study.Select only those experts who pass the identified criteria for further study.

ExampleA study of expert research engineers starts with an exploration of who other engineers look up to and who are most valued by their employers. The result determines that 'expert' can be defined as only those who have been awarded ten or more patents and who have at least twelve years of experience.

DiscussionOpinions of experts are more easily respected by other people. Studies that report expert opinion are likely to benefit from a reflected respect and so be more credible at least with an audience who unquestionably accepts those people as experts.A key part of the reported study may be in establishing the expertise of the people in the study.It is not uncommon for 'experts' to be selected on relatively simple criteria, for example where a professor from a local college is assumed to be able to pronounce authoritatively on their subject. In practice, they may be low down on the national

10

scale of professorial expertise and other academics may quickly question what they say.In this way 'experts' are popular with TV interviewers as they often sound like they know what they are talking about and are seldom questioned.Experts are sometimes the subject of study, for example where the differences between their views is used to highlight uncertainties in their field of expertise.

3. Diversity Sampling – seeking variation with a wide net

UseUse when you are studying diversity and seeking a wide range of views or different subjects.

MethodRather than constrain the targeting to limited groups and areas, spread the net as wide as possible to gain a wide range of subjects and views.

ExampleA study may seek to identify all ethnic groups within a city and thus may be made across a number of different parts of the city and includes questioning of each

DiscussionSome studies are broad rather then narrow and hence need to identify the widest possible set of subjects. Narrow studies seek to eliminate variation. Diversity studies seek to discover and understand variation.The final sample size in such studies may not be determined, although a minimum size may be identified. More critical is sampling across a wide area of the population in order to maximize the chance of identifying theNarrow studies may be preceded by wide, heterogeneous studies that help a second-stage study ofDiversity sampling is also known as Heterogeneity sampling.

2. Probability Sampling – involves randomization1.1 Simple Random Sampling

- most basic design- Table of Random Numbers and Lottery

1.2 Systematic Sampling with a Random Start- selection of every Kth case

UseUse when it is difficult to identify items using a simple random sampling method (with random numbers).Use when it is easier to select every nth item.

MethodIdentify your sample size, n. Divide the total number of items in the population, N, by n. Round the decimal down. This gives you your interval, k.Thus for a population of 2000 and a sample of 100, k = 2000/100 = 20.Put the population into a sequential order, ensuring the attribute being studied is randomly distributed.Select a random number, x, between 1 and k.The first sampled item is the x-th. Then select every k-th item.Thus if k is 20 and x is 12, select the 12th item, then the 32nd, then the 52nd and so on.In brief: select every nth item, starting with a random one.

ExampleA study of people going to night-clubs first determines that there are about 250-300 people in the club (due to fire regulations). A sample size of 30 is selected, giving an interval of 300/30 = 10. A random number between 1 and 10 is generated and comes up with 7. Starting with the 7th person to enter the club, every 10th person is given a brief interview.Other precautions are taken to neutralize any impact on the study of what time of night people people enter the club.

DiscussionThis method only works if you can sort the items being studied into a sequence in which you can ensure the studied attribute is random.It gives a handy method when a random number would be difficult to apply or when counting every nth item is simply easier. In the example above, if sequential random

11

numbers were used and the first random number was 250, then you would have to wait for the 250th person for your first item.Systematic sampling is also called systematic random sampling.

1.3 Stratified Random Sampling.- divides population into homogenous subsets- to enhance representativeness

UseUse it when there are smaller sub-groups that are to be investigated.Use it when you want to achieve greater statistical significance in a smaller sample.Use it to reduce standard error.

MethodDivide the population up into a set of smaller non-overlapping sub-groups (strata), then do a simple random sample in each sub-group.Strata can be natural groupings, such as age ranges or ethnic origins.

Example

A high school student who is studying year-ten attitudes in the school uses registration tuition classes as strata and studies a random selection of students from each of these classes.

In a company there are more men than women, but it is required to have each group equally represented. Two strata are thus created, of men and women, with an equal number in each.

DiscussionStratification aims to reduce standard error by providing some control over variance. If you know that there are groups that must be included, for example men and women, then you can deliberately sample these in a due proportion.Proportionate stratified sampling takes the same proportion (sample fraction) from each stratum.Disproportionate stratified sampling takes a different proportion from different strata. This may be done to ensure minorities are adequately covered. If you do this, and want to make an estimate about the population, you will have to weight within-group estimates using the sampling fraction.If the groups are homogeneous (ie. have the same proportions of each attribute), and hence within-group variation is lower than the population, then stratified random sampling will give a statistically more accurate result than simple random sampling.Stratified sampling is sometimes called quota sampling or stratified random sampling.

1.4 Cluster Sampling- successive random of units

UseUse when the studied population is spread across a wide area such that simple random sampling would be difficult to implement in accessing the selected sample.

MethodDivide the population up into a set of different coherent areas.Randomly select areas to assess.Access all subjects in the selected areas. If you cannot do this, select a significant random sample and use the same selection rules in each cluster.

ExampleIn a study of the opinions of homeless across a country, rather than study a few homeless people in all towns, a number of towns are selected and a significant number of homeless people are interviewed in each one.

DiscussionSometimes the biggest problem with sampling is being able to reach your

targets, and having them are spread out over a large geographic area is a common experience.

Even when you have selected a cluster, you are unlikely to be able to access everyone in that cluster (you are unlikely, for example, to be able to interview everyone in a selected town). The practical answer is to select a significant and similar sample in each cluster. For example if you are going to interview people in clothes shops, you should do this at the same time on the same weekday in

12

each cluster (you would, after all, likely get different results interviewing 9am Monday morning from if you did it on Saturday afternoon).

Cluster sampling may be combined with other forms of sampling, for example proportionate quota sampling, to ensure sub-groups are fully represented.

A risk with cluster sampling is that some geographic areas can have different characteristics, for example affluence or political bias.Cluster sampling is also called area sampling.

1.5 Multistage Sampling

DATA COLLECTION

Data – pieces of info obtained in the study

Types of Data: 1. Qualitative – info collected in narrative form (nonnumerical) form2. Quantitative – info collected in numerical form 3. Primary – first-hand reports of facts, findings or events, originally prepared by the

researcher.4. Secondary – data from a study prepared by someone other than the original researcher.

3 Approaches to Data gathering1. Self - Reports2. Observation3. Biophysiologic measures

1. Self-report – data are collected by means of an oral interview or written questionnaire.

Unstructured Interviewsa. Complete unstructured interviews

- conversational discussions on the topic of interestb. Focused (semi-structured) interviews

- uses broad topic guide, ensuring that all questions are covered.c. Focused Group interviews

- interviews with small group of people (5-15).- interviewer guides the discussion according to topic guide- advantage: efficient and can generate dialogues

d. Life histories- encourages respondents to narrate their life experiences

e. Diaries- respondents are asked to maintain daily records about some aspects of their lives

Structured Interviews – usually employs a formal instrument, a questionnaire / interview schedule

Closed-ended questions- offer respondents fixed alternatives to choose

Open-ended questions- permits respondents to respond in their own words

Likert Scale – declarative statements that express a viewpoint on a topic. - Respondents asked to indicate the degree to which they agree or

disagree with the opinion expressed by the statement.- an ordered, one-dimensional scale from which respondents choose one option that best aligns with their view.

 

5-point traditional Likert scale:  Strongly Tend to Neither Tend to Strongly

13

agree agree

agreenor

disagree

disagree

disagree

I like going to Chinese restaurants [  ] [  ] [  ] [  ] [  ]

 

5-point Likert-type scale, not all labeled:  Good   Neutral   Bad

When I think about Chinese restaurants I feel [  ] [  ] [  ] [  ] [  ]

 

6-point Likert-type scale:

  Never Infrequently Infrequently Sometimes

Frequently

Always

I feel happy when entering a Chinese Restaurant

O O O O O O

 Sematic Differential- respondents are asked to rate on bipolar adjectives, good/bad, effective/ineffective

Visual Analog Scale (VAS)- measures subjective experiences such as pain, fatigue, and dyspnea- a straight line, the end anchors o which are labeled as the extreme limits of the sensation or feeling being measured.

Vignettes – brief descriptions of events or situations to which respondents are asked to react, can either be factual or fictitious

Examples: How would you recommend handling this situation?On the nine-point scale below, rate how well you believe the nurse handled the situation?

Q Sorts- set of cards on which words or statements are written.

2. Observational Methods

Participant Observation – researcher observes and participates in the functioning of the group or institution under study

3. Biophysiologic Measures- assesses clinical variables- accurate and precise measurement (compared with psychological measures such as self-reports on anxiety, pain, etc.

- objective in nature

In vivo measures – performed directly within or on living organisms.Ex: blood pressure, body temp.

In vitro measures – performed outside the organism’s bodyEx: Laboratory tests, cytologic measures

VALIDITY – degree to which an instrument measures what it is suppose to be measuring

14

*A measuring device that is unreliable cannot be possibly valid. An instrument cannot validly measure an attribute if it is erratic and accurate.

*An instrument can, however, be reliable without being valid. Suppose, we wanted to assess patient’s anxiety by measuring the circumference of their wrists. We could obtain highly accurate and precise measurements of wrists circumferences, but such measures would not be valid indicators of anxiety.

Face Validity – whether an instrument looks as though it is measuring the appropriate construct

3 Kinds1. Content Validity

– based on experts judgement.- sometimes called logical or rational validity, is the estimate of how much a

measure represents every single element of a construct.

For example, an educational test with strong content validity will represent the subjects actually taught to students, rather than asking unrelated questions. Content validity is often seen as a prerequisite to criterion validity, because it is a good indicator of whether the desired trait is measured. If elements of the test are irrelevant to the main construct, then they are measuring something else completely, creating potential bias.

In addition, criterion validity derives quantitative correlations from test scores.Content validity is qualitative in nature, and asks whether a specific element enhances or detracts from a test or research program.

HOW IS CONTENT VALIDITY MEASURED?Content validity is related to face validity, but differs wildly in how it is evaluated.

Face validity requires a personal judgment, such as asking participants whether they thought that a test was well constructed and useful. Content validity arrives at the same answers, but uses an approached based in statistics, ensuring that it is regarded as a strong type of validity.

For surveys and tests, each question is given to a panel of expert analysts, and they rate it. They give their opinion about whether the question is essential, useful or irrelevant to measuring the construct under study.Their results are statistically analyzed and the test modified to improve the rational validity.

AN EXAMPLE OF LOW CONTENT VALIDITYLet us look at an example from employment, where content validity is often used.A school wants to hire a new science teacher, and a panel of governors begins to look through the various candidates. They draw up a shortlist and then set a test, picking the candidate with the best score. Sadly, he proves to be an extremely poor science teacher.After looking at the test, the education board begins to see where they went wrong. The vast majority of the questions were about physics so, of course, the school found the most talented physics teacher. However, this particular job expected the science teacher to teach biology, chemistry and psychology. The content validity of test was poor and did not fully represent the construct of ‘being a good science teacher.Suitably embarrassed, the school redesigned the test and submitted it to a panel of educational experts. After asking the candidates to sit the revised test, the school found another teacher, and she proved to be an excellent and well-rounded science teacher. This test had a much higher rational validity and fully represented every element of the construct.

2. Criterion-related validity - researcher seeks to establish relationship between the scores on an

instrument and some external criterion. The instrument, whatever it is measuring is said to be valid, if its scores correspond strongly with scores on criterion.

15

Ex: A researcher wanted to find out the intelligence of college freshman by using their College Aptitude test. In order to determine whether it is valid, a researcher may use an outside criterion such as High School General Average.

Predictive Validity - instruments ability to differentiate between people’s performances or behaviors on future direction.- When a school of nursing correlates students’ incoming high school grades with their subsequent grade-point averages, the predictive validity of high school grades for nursing school performance is being evaluated.

Concurrent Validity- instruments ability to distinguish among people who differ in their

present status on some criterion. For example, a psychological test to differentiate between patients in mental institutions who could and could not be released could be correlated with current behavioral ratings of health care personnel.

*the difference between concurrent and predictive validity is the difference in the timing of obtaining measurements on a criterion.

3. Construct-related validity- adequacy of an instrument in measuring the construct of interest.- Construct validity defines how well a test or experiment measures up to its claims. It refers to whether the operational definition of a variable actually reflects the true theoretical meaning of a concept.

”What is this instrument really measuring?”

a. known-groups techniqueb. factor analysis

Sensitivity – ability of a measure to identify a “case” correctly, that is, to screen in or diagnosis a condition correctly. Measured by “true positives”.

Specificity – measures ability to identify noncases correctly, that is, to screen out those without the condition. Yields “true negatives”.

Credibility – confidence in the truth of the data.*Prolonged engagement – investment of sufficient time in data collection activities to

have an in-depth understanding of the culture, language or views of the group under study and to test for misinformation.

*Persistent observation – refers to the researcher’s focus on the aspects of situation that are relevant to the phenomenon being studied

*Triangulation – uses multiple referents to draw conclusion

Example of Triangulation: A researcher studies the experiences of nursing students with disabilities. The study involved data triangulation(interviews were conducted with students with physical or auditory impairments and nursing faculty members, patients, and fellow students).

Dependability – data stability over time and over conditions

Confirmability – objectivity or neutrality of the data

Reliability – consistency and accuracy of data

Methods of Determining Reliability:1. Test-retest – administration of the same test twice, responses may change over time.

uses reliability coefficient, objectively determines how small the differences are.2. Equivalent Form Method – administering the 2 different exams

uses coefficient alpha (Cronbach’s alpha). The normal range is between .00 and +1.00. The higher the reliability coefficient, the more accurate the measure.

16

3. Internal Consistency method – administering the exam once*Split-half technique – items comprising a scale is split into 2 groups.

Data Processing Steps:

1. Editing – to determine incomplete responses, omissions, inaccurate info2. Coding – conversion of all possible responses into numerical value

Ex: 1 – male, 2 – female3. Encoding4. Creation of Data Files5. Tabulation

DATA ANALYSIS and INTERPRETATIONData Analysis

Descriptive Analysis – used to describe and synthesize dataInferential Analysis – to test hypothesis, draw conclusions about population

Scales of Measurement:1. Nominal Scale – no numbers involved. E.g. gender, blood type2. Ordinal Scale – order/rank, e.g. degree of burn, stage of cancer3. Interval Scale – no zero value, e.g. BUN level, temp. of human body

4. Ratio Scale – with zero point value, e.g. No. of Children,

Methods of Analyzing the Data:

Frequency Distribution – method of imposing order on raw data, numeric values are ordered from lowest to highest

Positive skew – longer tail is pointed towards the rightNegative skew - longer tail is pointed towards the left

Measures of Central TendencyMean - averageMedian – midpoint valueMode – frequently occurring

Variability – degree to which values on a set of scores are widely different or dispersed*Range – highest score minus the lowest score distribution

Ex: AIDS knowledge test scores, range is 15 (30-15) + 1*Standard Deviation – most widely used to determine variability

Bivariate Descriptive Statistics- describe relationship between variables

Correlation Coefficient – describes intensity and direction of relationship*positive- .00 - +1.00*negative - -1.00 - .00

Product-moment correlation coefficient (Pearson’s r)- computed with interval and ratio measures

Spearman’s rank-order correlation (Spearman’s rho)- for ordinal measures

T-test – appropriate for testing the statistical significance of a difference between means of two groups.

Analysis of Variance (ANOVA) – used to test mean group differences of three or more groups

- can be used to test the effect of two (or more) independent variable.

Multiple Comparison Procedures (post hoc tests)

17

- isolates the differences between group means that are responsible for rejecting the overall ANOVA null hypothesis.

Chi-Squared (X2) Test – used to test hypothesis about the proportion of cases that fall into various categories.

- Add the differences bet. observed frequencies and expected frequencies (the frequencies that would be expected if there were no relationship between two variables).

Correlation Coefficient – describe the direction and magnitude of a relationship between two variables, and range from -1.00 (perfect negative correlation) through .00 to +1.00 (pefect positive correlation

*Pearson’s r – used with interval-or ratio-level variables

Multivariate Statistical AnalysisMultiple Regression

- allows to explain or predict a dependent variable with multiple independent variables. - the dependent variables are interval-or ratio-level variables.- Independent variables (predictor variables in multiple regression) are either interval – or ratio – level variables or dichotomous nominal level

Multiple Correlation Coefficient (R)

Analysis of Covariance (ANCOVA)- combines ANOVA and multiple regression- used to control confounding variables statistically – that is to “equalize” groups being compared. - to test the difference between the means of 2 groups, while controlling for one (1) covariate.

Nurture your mind with great thoughts, for you will never go any higher than you think.

mdgolveo

18