Midterm Flashcards
Abstract
a brief description of the study placed at the beginning of the article o Concise summary of the study o Usually less than 300 words o Should provide the following: Research problem Results / findings Main conclusions Recommendations
Best source of evidence
Systematic Reviews- (level 1) the strongest evidence comes from careful synthesis of multiple studies
Bias
a distortion or influence that results in an error in inference
Bivariate descriptive statistics: Cross tabulation
frequencies of two variables are analyzed against each other
Bivariate descriptive statistics: Correlation
association between variables
Bivariate descriptive statistics: Correlation coefficient
describes intensity and direction of a relationship
correlation coefficient range from −1.00 through .00 to +1.00
Bivariate descriptive statistics: Positive relationship
relationship between two variables in which high values on one variable tend to be associated with high values on the other
Bivariate descriptive statistics: Negative (inverse) relationship
variables are inversely related, higher values on one variable are associated with lower values in the second
Bivariate descriptive statistics: Pearson’s r
most widely used correlation statistic
computed with continuous measures
Need to have interval or ratio level variables
Bivariate descriptive statistics: Spearman’s rho
correlations between variables measured on an ordinal scale
Calculating median
o 1, 5, 2, 8, 7
o sort numbers: 1, 2, 5, 7, 8
o Median is 5
o If even numbers, median is the arithmetic average (mean) of the 2 numbers
Consecutive sampling
o Recruitment of all people from an accessible population over a specific time interval or for a specified sample size
o Best possible choice when there is “rolling enrollment” into an accessible population
Convenience sampling
selecting the most conveniently available people as participants
o People might be atypical of the population (risk of bias)
o Weakest form of sampling, also most commonly used sampling method
Credibility
A criterion for evaluating integrity and trustworthiness in qualitative studies, referring to confidence in the truth of the data; analogous to internal validity in quantitative research
Critical theory research
o Critique of society
o Envisions new possibilities
o Action oriented
o Inspire change
Data saturation
involves sampling until no new information is obtained and redundancy is achieved
o Participants are insightful- saturation achieved with a small sample
o A larger sample is needed with maximum variation sampling
Debriefing
Communication with study participants after participation is complete regarding aspects of the study (explaining the purpose of the study more fully)
Descriptive statistics
o Used to synthesize/summarize/ describe data. o Parameters: Characteristics of a population o 3 Characteristics: Shape of distribution of values Central tendency Variability
Frequency distribution positive skew
o Asymmetrical, the peak is off center, and one tail is other than the other.
o When one tail points to the RIGHT the distribution has a positive skew.
Inference
a conclusion drawn from the study evidence using logical reasoning and taking into account the methods used to generate the evidence
Levels of measurement: Nominal
Lowest level Organized in categories Cannot be compared Exclusive Ex: gender, ethnicity, marital status, diagnosis
Levels of measurement: Ordinal
Organized in categories Can be compared (ranked) Rank higher or lower/better or worse Exclusive Unequal intervals Ex: mild, moderate, severe pain OR…no exercise, some exercise, heavy exercise
Levels of measurement: Interval
Equal numerical distances between intervals Exclusive Continuum value LACKS A ZERO POINT! Ex: temperature
Levels of measurement: Ratio
Highest level Exclusive Ordered Equally spaced interval Continuum value DOES HAVE ABSOLUTE ZERO Ex: weight, height, volume
Measures of central tendency: mode
Value that occurs most frequently
Used with nominal data
Bimodal distribution
Multimodal distribution
Measures of central tendency: median
Exact center of frequency distribution
Think midpoint
Used with ordinal data
Not affected by extreme scores
Measures of central tendency: mean
Sum of the scores divided by the number of scores being summed
Used with interval or nominal data
Member check
a method of validating the credibility of qualitative data through debriefings and discussions with informants
Outliers
a data point that differs significantly from other observations
PICO question
o P- population Age Gender Ethnicity With certain disorder (diabetes) o I-interventions Exposure to disease Risk behavior o C-comparison No disease Placebo or no intervention or therapy Absence of risk factor o O-outcome Risk of disease Accuracy of diagnosis Rate of occurrence of adverse outcome
Prolonged engagement
the investment of sufficient time collecting data to have an in-depth understanding of the culture, language, or views of the people or group under study
o Tests for misinformation
o Ensures saturation of important categories
o Important for building trust with informants
Purposive sampling
o Using researchers’ knowledge about the population to handpick sample members
o Leads to bias but useful approach when researchers want a sample of experts
Qualitative sampling methods: Convenience (think volunteer)
Volunteer samples come forward and identify themselves.
Efficient, economical, and convenient
Not preferred
Qualitative sampling methods: Snowball (think network)
Asking informants to make referrals
Restrictive to small network
Lower quality in sample
Qualitative sampling methods: Purposive (think purposeful)
Several different types Deliberately choose (on purpose) the cases that will contribute to the study. • Maximum variation sampling • Extreme (deviant) case sampling • Typical case sampling • Criterion sampling
Qualitative sampling methods: Theoretical
Used in grounded theory
• 20-30 sample size
Based on the development of the theory
Quasi-experimental design advantages
o More practical for a natural setting
o Useful for people who are not always willing to be randomized
Has generalizability of the results
Research findings
are meant to reflect “truth in the real world”
o Inference involves drawing conclusions based on limited information and using logical reasoning
valid to the extent that the researchers have made good decisions in selecting proxies and have controlled sources of bias
Research hypothesis
Predicted answers to research questions. State the expected relationship between the IV and DV
o Offer direction
o Suggest explanations
o Promotes critical thinking
o Directional Hypothesis-Expected direction of the relationship
o Nondirectional Hypothesis-Does not stipulate direction of the relationship
o Null Hypothesis-Absence of relationship. H01
Research problem statement elements
o 1.) Problem identification o 2.) Background o 3.) Scope of problem o 4.) Consequences of the problem o 5.) Knowledge gaps o 6.) Proposed solution
Retrospective design
Outcome (effect) observed is linked to potential cause from the past
Sample size
the number of people who participate in a study; an important factor in the power of the analysis and in statistical conclusion validity in quantitative research
Semi structured interviews (or focused)
o Researchers have a list of list of topics or broad questions that must be covered in an interview
o Use a topic guide
Simple random sampling
o most basic probability sampling
o establish a sampling frame (a list of population elements)
o 1. Elements in a sampling frame are numbered
o 2. A table of random numbers or an online randomizer is used to draw a random sample of the desired size.
o Samples selected randomly are not likely to be biased.
o There is no guarantee of a representative sample
Statistical test
An analytic tool that estimates the probability that obtained results from a sample reflect the population values
Cross-sectional
Data collected at one point in time
Economical
Pose problems for inferring changes over time
Longitudinal
Data collected multiple times over an extended period
Concern – attrition (losing participants over time)
Type II error
o accepting the null hypothesis when it is false
o false-negative conclusion
Types of knowing
o Traditions = customs o Authority = expertise o Borrowing = other fields o Trial and Error = unsure how to approach o Personal Experience = gaining knowledge o Role Modeling =Imitating o Intuition = gut feeling o Reasoning = critical thinking/logic
Types of measurement: Direct
Measurement of concrete items
Weight, height
Types of measurement: Indirect
Captures elements of abstract concept
Coping, anxiety
Types of nonprobability sampling: Convenience sampling
- Selection of the most readily available persons as participants
- BIAS!
Types of nonprobability sampling: Quota sampling
- A nonrandom sampling method in which “quotas” for certain subgroups, based on sample characteristics, are established to increase the representativeness of the sample
- Researchers identify population strata and figure out how many people are needed from each stratum
- Quota sampling is similar to convenience sampling: Participants are a convenience sample from each stratum
Types of nonprobability sampling: Consecutive sampling
- Recruitment of all people from an accessible population over a specific time interval or for a specified sample size
- Best possible choice when there is “rolling enrollment” into an accessible population
Types of nonprobability sampling: Purposive sampling
- Using researchers’ knowledge about the population to handpick sample members
- Leads to bias but useful approach when researchers want a sample of experts
Types of qualitative research: Ethnography
Description and interpretation of culture & cultural behavior
Extensive fieldwork (time consuming)
Purpose: seek to learn from rather than study
-emic & -etic perspectives
o Information sought:
Cultural behavior
Cultural artifacts
Cultural speech
o Participant observation
Strategy used by ethnographers
Make observations while participating in activities
Types of qualitative research: Phenomenology
Main data source is in-depth conversations Key themes o Two main types: Descriptive Interpretative
Types of qualitative research: Descriptive
(What do we know as persons?)
Depicts “things” as people experience them:
Eclectic design
In-depth collection of qualitative data
No traditional qualitative roots
Content analysis for data interpretation
• Organizing data in key concepts/themes
Types of qualitative research: Interpretative
Rely on in depth interviews
• May augment understandings
Lived experience and interpretive process
Goals:
• To enter another’s word & discover the understandings.
Also referred to as Hermeneutics = understanding
• Bracketing does not occur in interpretive phenomenology
Types of qualitative research: Grounded theory
Develops middle range theories
Identify main problem and offer resolution
Constant comparison is used
• New data collected compared ongoing to previous data collected
In-depth interviews & participant interviews for data collection
Types of qualitative research: Historical
Systematic collection & critical evaluation of data related to past events/occurrences
Usually interpretive-describe what, how, & why
Primarily relies on written records:
• Review data from diaries, letters, newspapers, medical documents
o nonwritten data: photos & films
• interview people who experienced past events
Types of qualitative research: case studies
Single or Small number of entities with in-depth review
Understand the “why” rather than the “what”
Strength/Weakness
• Small sample and get most in-depth data
Types of qualitative research: narrative analysis
The story is the form of inquiry and understanding
Types of qualitative research: critical theory
o Similar to critical theory
Focuses specifically on gender domination/discrimination
Goal: facilitate change to end women’s unequal social position
Types of qualitative research: feminist research
Production of knowledge can influence power
• Also focuses on action, empowerment, and raising awareness
Types of quantitative research: experimental
Randomized Controlled Trials (RCT)
• “True Experiment”
• Level 1 (pg 23 Fig 2.1)
• Gold standard
• Test the effects of an intervention AND random groups assigned
Pretest-Posttest design
• Collect pretest data (baseline) – perform intervention – Collect posttest data (outcome)
Crossover design
• Exposing people to more than one treatment
Types of quantitative research: quasi-experimental
nonequivalent control group pretest-posttest design
• Level II
• Weaker because no randomization
Trials without randomization
Similar to pretest-posttest experimental design without randomized groups
How to identify:
• Testing intervention
o Comparison group instead of control group
No mention of random group assignment
Time-series Design
Types of quantitative research: correlational
Level III Examine relationship between variables Non-experimental study Correlations can be detected through statistical analyses Does not prove causation Specific correlational designs: • Prospective design (cohort design) o Start with a presumed cause • Retrospective design o Outcome (effect) observed is linked to potential cause from the past • Case-Control design o Review cases that meet certain criteria and identify controls
Types of quantitative research: descriptive
Level III
Observe, describe, document situation
Types of quantitative research: time dimensional (cross sectional/longitudinal)
o Cross-sectional
Data collected at one point in time
Economical
Pose problems for inferring changes over time
o Longitudinal
Data collected multiple times over an extended period
Concern – attrition (losing participants over time)
BSN
Locate and critically appraise studies, evidence-based guidelines, protocols, and policies with assistance.
Use best research evidence in practice with guidance.
Assist with problem identification and data collection.
MSN
Critically appraise and synthesize studies to develop and review protocols for practice.
Implement best research evidence in practice.
Collaborate in research projects and provide clinical expertise for research.
DNP
Critically appraise studies and develop and refine protocols and policies for practice.
Participate in evidence-based guideline development.
Implement, evaluate, and revise as needed protocols, polices, and evidence-based guidelines in practice.
PhD
Major role in conducting independent research and contributing to the empirical knowledge generated in a selected area of study.
Obtain initial funding for research.
Coordinate research teams of BSN, MSN, and DNP nurses.
