Research Methods PHU4410
Median calculation
(1) Rank the data set in increasing order (or decreasing order) (2) Find the middle term **Note: If the number of observations is even, then the _____ is given by the average of the values of the two middle terms
What are index terms?
(also called Subjects or Subject Headings) are single words or brief phrases that describe the document's content, but they are chosen from a pre-existing list (also called a controlled vocabulary).
What does PICOS stand for?
*P*opulation / *P*atient *I*ntervention *C*omparison *O*utcomes *S*tudy design
How to increase response rate?
- Maximize rewards of participation - Minimize costs of participation - Maximize trust - Personalised questions or personal touch (name on envelope) - stamped envelopes - shorter questionnaires - good communication - assure confidentiality
what is a mixed methods study?
-Combination of quantitative and qualitative -Looks at both statistics and experience/themes -Requires a lot of work cuz you have to use both; not always feasible
Is an animal welfare body a statutory requirement?
Yes
. Descriptive statistics focus on two particular properties of a dataset: a) Central tendency of a distribution and dispersion b) Central tendency of a distribtution and p values c) Dispersion and p values d) Dispersion and control samples e) Type 1 and type 2 error
a
A list of 5 pulse rates is: 70, 64, 80, 74, 92. What is the median for this list? a) 74 b) 76 c) 77 d) 80 e) 92
a
If a sample size is 5, which of the following provides more details in the tails of the distribution? a) Box plots b) Histograms c) Pie-charts d) SPSS datasets e) Venn diagrams
a
The 3Rs are quite simple in concept but somewhat difficult to implement consistently in practice. Experience is important. In the following section, which of the following is not correctly matched. a) Reduction: Permitting animal experimentation solely for the purpose of drug discovery studies b) Replacement: Using in-vitro (e.g. cell or tissue culture) instead of in-vivo (whole organism) techniques c) Refinement: Using analgesics or drugs to minimise pain or discomfort d) Refinement: housing wild animals in more spacious and natural settings e) All are correctly matched
a
he HPRA is the competent authority under Directive 2010/63/EU for the protection of animals used for scientific purposes, in respect of each of the below stated function, what is the function of the HPRA? a) Development of related policy and legislative developments b) The authorisation of projects and individuals c) Ensuring compliance with the 3R principles (replacement, refinement and reduction) by carrying out regular inspections. d) Monitoring the adequacy od record keeping at establishments in line with the requirements of the Directive and reporting on specified Statistical parameters to the EU Commission e) Liaising with the competent authorities under Directive 2010/63/EU from other EU Member States regarding the sharing of information and best practices relating to the 3Rs.
a
MCQ Q: When designing a protocol for a qualitative study, which of the following guidance should be followed when writing the protocol? a) PICO b) SPIDER c) SPICE d) PROSPERO e) COREQ
a (think pic said c tho?)
What is EMBASE?
a biomedical and pharmacological bibliographic database of published literature
What is a confounding variable?
a factor other than the independent variable that might produce an effect in an experiment
What is qualitative research?
a systematic, subjective approach to research that is used to describe life experiences and situations and give them meaning.
What is an animal welfare body responsible for?
advising scientists and staff on matters related to welfare, 3Rs, rehoming schemes, ensuring best practice, monitoring programs for sharing of organs and tissues of euthanised animals, giving records to HPRA of any advice provided
What is a population in statistics terms?
aggregate of things (e.g. objects, events) or creatures
Who are non-essential animal research ethics committee members that can enhace the value of asssessment?
an ethicist, a patient representative, a specialist in animal species being used, additional animal technicians?
What is a confounder?
an extraneous variable whose presence affects the variables being studied so that the results do not reflect the actual relationship between the variables under study i.e. independently associated with both variables being studied ex: increased rate of pancreatic cancer among coffee drinks. Coffee drinkers are more likely to smoke must be a parallel pathway to the same result must be linked to both exposure and disease and not an intermediate in the cause-effect relationship
Define data
an information about (or deriving from) a study including details of methods, participants, setting, context, interventions, outcomes, results, publications and investigators
What is JBI (Joanna Briggs Institute)?
an international research organisation which develops and delivers evidence-based information, software, education and training designed to improve healthcare practice and health outcomes
Time to event data
analyses time until event occurs, but where not all individuals in the study experience the event
What is the definition of a survey?
any activity that collects information in an organised and methodological manner about characteristics of interest from some or all units of a population using well-defined concepts, methods and procedures and compiles such information into useful summary form
What are supporting documents with examples?
any document related to how research was done e.g. letter of invitation, consent form, audit data collection form, demographic matrix
What is generalizability in research?
application of the results to overall population of interest
A researcher wants to study the effect of 3 different drugs on 3 independent groups of rats (10 rats per group). What would the alternative hypothesis be? a) All group means are equal b) At least one group mean is different c) All group standard deviations are equal d) All group standard deviations are different e) The Spearman rank-order correlation coefficient should be 0.05
b
Counts and rates data
based on number of events each individual experiences
What are the ARRIVE essential 10?
basic minimum that must be included in any manuscript describing animal research
why are blinding and allocation concealment different?
blinding after they are allocated, allocation before
Assuming all other relevant factors remain constant, the power of a statistical test: a) Decreases as the size of the sample increases b) Increases as the significance level decreases c) Increases as the variability of the observations decreases d) Is the chance of not rejecting the null hypothesis when it is true e) Is the chance of rejecting the null hypothesis when it is true
c
If a series of values consists of 20 numbers, then, for finding the mediain, we ordered the series ascending and we use: a) The 10th value in the ordered series b) The 11th value in the ordered series c) The mean between the 10th and 11th values d) The mean between the 11th and 12th values e) The median cannot be calculated
c
What is the name of EMBASES controlled vocab? a. advanced search b. mesh c. emtree d. single citation matcher e. title/abstract
c
What kind of data is blood groups?
categorical, nominal
What is the purpose of methods in research?
describes how aim and objective will be assessed and allow reproducibility of the study
For a random effects model, what is the pooled result viewed as?
estimate refers to centre of distribution of treatment effects
What unit of analysis is used in quantitative research?
numerical
What must be assessed to determine the quality of a questionnaire?
reliability and internal consistency
What are disadvantages of a qualitative focus group?
time consuming, expensive, confidentiality concerns, dominant group members may affect input from other participants, potential for response bias
Calculate the risk ratio; if the risk of exacerbation in intervention group was 3 in 63 (RISK = 0.05) , and the risk of exacerbation in the control group was 10 in 64 (RISK = 0.16). RR =
0.31
Moderate Correlation what values for r?
0.5 > r < 0.8 and -0.5 < r > -0.8 (i.e. r between 0.5 and 0.8)
What should a protocol contain?
1. Objectives and purpose of study. 2. Investigators information - address, name, etc. 3. Criteria for pt. selection and # of pts. 4. Description of study design. 5. Method of determining dosage - max dose/duration. 6. Observations and measurements. 7. Description of clinical procedures.
What circumstances would require inferential statistics?
1. You have a complete list of participants in your population of interest 2. You are able to draw a truly random sample from the population of interest 3. You have established that the size of your truly random sample is sufficient to enable you to draw inferences about your findings
Steps for choosing the right statistical test?
1. decide independent/predictor/explanatory variable and the dependent/outcome variable first 2. is the data normally distributed - parametric it is, non-parametric, it is not 3. how many predictor/independent variables are there 3. determine test
What are the key stages of a systematic review?
1. formulate question 2. develop protocol 3. conduct review - searches, eligibility criteria, data extraction, critical appraisal, analyse and interpret results, write up review 4. update or improving review
How can selective reporting occur?
1. selective omission of outcomes 2. selective choice of data for an outcome 3. selective reporting of analyses using the same data 4. selective reporting of subsets of data 5. selective underreporting of data
What items must be assessed in a risk of bias assessment?
1. sequence generation (randomisation) 2. allocation sequence concealment 3. blinding of participants/personnel 4. blinding of outcome assessors 5. incomplete outcome data 6. selective outcome reporting
How many studies must you use for a meta-analyses (minimum)?
2
Determing what test to use; numerical data -> two groups -> independent
2 sample t-test
How many times should an experiment be repeated to confirm results?
3
What is the minimum sample size for a histogram?
30
What is the minimum sample size for a box plot?
5
Why don't we just use three t-tests instead of post-hoc?
5% risk each time of making a Type I error, therefore three times would give us 15% error
What is the minimum number of people composing of an animal research ethics committee?
6
How many studies in a meta-analysis do you need to perform a meta-regression?
>10
Which should you use to compare numerical data between two groups? A. Box plot B. Histogram C. Scatter plot
A
How many reviewers should perform a risk of bias assessment?
2
What value of I2 in corresponds to small heterogeneity?
25%
What is an ideal sample power %?
80%
What is a Type I error?
A Type I error occurs when the null hypothesis is rejected when it is true, and significant difference when there is no difference between the populations In plain terms...this means, that the researchers conclude that significant results exist in a study, when really, they don't.
what is an inference?
A conclusion reached on the basis of evidence and reasoning
What is COCHRANE
A database of systematic reviews of primary research in human health care and policy
What is a focus group?
A group interviewing or discussion technique which is used to enhance understanding and gain in-depth knowledge by gauging consumer feedback
I have three groups of subjects, the subjects are independent of eachother and the data is numerical, what test should I use
ANOVA
What is a project authorisation in animal research?
Allow to perform scientific procedures on animals
What is a double blind experiment?
An experiment in which both the individuals doing the testing and being tested are unaware of the hypothesis and who is in either the experimental or control group
Detection bias
Arises from educated professionals using their knowledge in an inconsistent way by searching for an outcome disproportionately in certain populations
What bias does intention to treat prevent?
Attrition
Performance bias
Bias that occurs due to systematic differences in care between treatment groups or in exposure to factors other than the intervention being studied.
What can be used to reduce the likelihood of type I errors?
Bonferroni correction
Who must have an animal welfare body?
Breeders, suppliers and user establishments
How do you determine sample size?
By doing a power analysis
/What guideline for an randomised control trial
CONSORT
What are the roles of an individual who is authorised to carry out animal research?
Carry out procedures, act as project manager for authorisation and euthanisation
What must HPRA publish to the European Commission in terms of animal research?
Collect and publish statistical information annually on the use of animals?
What tools can you use for screening and selection of articles?
Covidence, Mendeley, ProQuest Refworks, Endnote
What are epidemiological studies?
Cross-sectional studies Case-control studies Cohort studies Longitudinal studies
What is discrete data?
Data can only take particular values. (Examples include number of apples or hair colour)
What are the two main types of observational studies?
Descriptive vs. Analytical
What does double coding do?
Enhance trustworthiness of a study
What are the requirements of a designated veterinarian in ethics committee?
Experience or specialist training in laboratory animal medicine and 3Rs, may be affiliated with establishment, species-specific training (for e.g. farm animal projects)
If cell count <5, do you use a Fisher's Exact Test or Chi-Square test for categorical data?
Fisher's Exact Test
What test should you use; you want to examine the association between gender and polypharmacy (+5 medicines) in a small dataset?
Fisher's Exact Test
What are the levels of evidence?
I- High Quality RCT's, Systematic Reviews of RTC's, Meta-Analysis II- Small-Scale RCT's, non-randomized studies with a control group III- Non-randomized studies with one group (no control group) IV- Descriptive Studies (single subject designs, case studies & surveys) V- Expert Opinion, Literature Reviews
What is step 1 of the scientific method?
Identify the problem
How should individual studies deal with missing outcome data?
Intention to treat analysis
What is statistical significance for p?
Less than 0.05
What is step 7 of the scientific method?
Make conclusions
What test would you use; determine if patient's had high blood pressure (yes or no) after one month before and after a medication?
McNemar's for paired data that can be categorised
What score does a downgraded RCT/upgraded observational study get with the GRADE approach?
Moderate quality rating
What is naturalistic research?
Occurs in the natural context/setting Minimal interference by the researcher Various degrees of observation/participation
What is a p-value?
Outcome of a statistical test Probability that the result obtained was due to chance. Generally a P-value < 0.05 (and sometimes < 0.01 or other values, depending on the trial design) indicated statistical significance. If P < 0.05 that means there is a < 5% probability that the result occurred by chance.
If there is selective omission of outcomes from report, what kind of bias is present?
Reporting Bias
If relative risk = 1
Risk in exposed is equal to risk in non-exposed (no association)
What is another term for relative risk?
Risk ratio
What is the Cochrane risk of bias tool called ?
RoB 2
What guideline for qualitative research
SRQR/COREQ
What guideline for an observational study
STROBE
Factors that affect the p value yielded by a t test include:
Sample size Amount of difference between the means Amount of variation
What does SMART goals stand for?
Specific, Measurable, Attainable, Realistic, Timely
What is information bias?
Systematic error that arises because the information collected about or from study subjects is erroneous
What is step 4 of the scientific method?
Test the hypothesis
Pearson correlation test
Tests for the strength of the association between two continuous variables, numerical, normal distribution
What is a dependent variable?
The dependent variable is the variable that is measured, what the scientist thinks will be affected.
What is assent?
The informal agreement obtained from a person who is unable to fully participate in the informed consent process, but who can provide a preference related to medical care
What is an experimental unit?
The smallest unit of analysis in the experiment from which data will be collected e.g. animal or object that is studied
What is the explicate order?
The way in which our subjective sensory systems perceive the world.
What kind of error is a false positive?
Type I error i.e. the one ur mom wished she got on her pregnancy test
What type of error is this; a test for a disease may report a negative result, when the patient is, in fact, infected.
Type II 'false negative'
Under what initiative were the ARRIVE guidelines developed?
UK National Centre for Replacement, Refinement and Reduction (NC3Rs)
McNemar test
Used to compare dichotomous (nominal) dependent paired categorical outcome variables; non-parametric
What is data saturation?
When no new insights/themes are emerging based on concomitant data analysis (AKA data repetition)
What is a control?
The variable that is kept constant, eliminates other factors affecting the outcome other than the independent variable.
What kind of post hoc test would you use; you have three treatments known to all reduce inflammation but you want to find the best treatment at reducing mean inflammation?
Tukey - pairwise comparisons of tests to eachother
What type of error in hypothesis testing is this; researcher concludes that cocaine causes hairloss when it doesn't actually as there is no significant significance between a bald coke user and one with hair
Type I error - conclude significant results when there isn't any "false positive"
What are characteristics of a good experiment?
Unbiased, powerful, simple, wide range of applicability, amenable to statistical analysis.
What is a designated veterinarians function in animal research ethics committee?
Uphold welfare, apply 3Rs and provide expertise on feasibility of the proposed project
How to prevent reporting bias?
Use multiple people to code the data. ... Have participants review your results. ... Verify with more data sources. ... Check for alternative explanations. ... Review findings with peers.
What is replacement in animal research?
Using a scientifically satisfactory method or testing strategy not using animals in the procedure is used e.g. cell culture, computer modelling.
What graph should be used to show overlap between antihypertensives and polypharmacy for categorical variables?
Venn Diagram
What are proximity operators?
WITHIN, NEAR, NEXT
What is a type II error?
When the null hypothesis is accepted when it is false. In plain terms...this means the researchers conclude that the study results are non-significant when in reality...the results were significant!
Do you need a project authorisation in animal research if you already have an individual authorisation?
Yes
Is dichotomous data the same as binary?
Yes
A researcher has carried out an experiment testing the apoptotic effect of a drug at different concentrations (10µM, 100 µM and 500 µM) on Caco-2 adenocarcinoma cells. The researcher wants to compare the mean cell count of the 3 treatment groups with a control group. Therefore, not all pairwise comparisons will be made. What kind of statistic test would be the best for the researcher to use? a) One-way ANOVA followed by Dunnet's post hoc test if ANOVA is significant b) One-way ANOVA followed by Tukey's post hoc test if ANOVA is significant c) One-way ANOVA followed by Games Howell post hoc test if ANOVA is not significant d) Two-sample independent t-test e) Two-way ANOVA
a
In weight loss programme, 50 participants were randomised to two groups. One group were instructed to eat lunch before 2pm and the other were instructed to have lunch after that. The weight loss for each patient was recorded. If the data is normally distributed, choose the most appropriate procedure to decide if there is a relationship between the time of day a person eats. a) Chi-squared test for independence b) Linear regression c) Paired-sample t-test d) One-way ANOVA e) Two-sample independent t-test
a
MCQ QUESTION FROM LAST YEAR: To avoid experiment bias, when neither the experimenter nor the participant is aware of which group the participant is in, this is a) Double blind b) Null hypothesis c) Placebo effect d) Random assignment e) Variable manipulation
a
Select which of the following statements is true: a) The probability of type 1 error is the chance of rejecting the null hypothesis when it is true b) The probability of type 1 error is the chance of not rejecting the null hypothesis when it is true c) The probability of type 2 error is the chance of rejecting the null hypothesis when it is false d) The probability of type 1 error is always greater than that of a type 2 error e) If we conduct 3 t -tests to compare 3 independent groups instead of an ANOVA, the probability to make a type 1 error is 5% overall
a
What is the definition of standard deviation? a) A quantity expressing the spread of dispersion of data from the mean b) Descriptive statistic which highlights the highest and lowest numbers within a dataset c) The commonest value in a dataset d) Range withing which most of the measurements lie e) A quantity expressing the spread or dispersion of data from the mod
a
Which non-parametric statistical test is used to assess the difference between 2 independent samples when the distribution is skewed? a) Mann-Whitney U test b) Wilcoxon sign-rank test c) Kruskal-Wallis test d) Spearman rank correlation e) Two-sample independent t-test
a
Which of the following best describes 'reproducibility' with regards to questionnaires? a) The extent to which the questionnaire produces repeatable findings b) The extent to which the questionnaire measures what it is intended to measure c) The applicability of the results to the overall population of interest d) A list of questions administered to survey participants e) The time and effort required of a respondent to complete a questionnair
a
Which of the following best describes a lapse as defined by Reason's Human Error Theory? a) A failure of expertise b) A lack of expertise c) A failure of attention d) A deliberate rule break e) A failure of memory
a
Which of the following is not intendent by the ARRIVE Guidelines; a) Be a guide for gaining ethical approval to facilitate publication b) Guide authors as to essential information to include in a publication and not be absolutely prescriptive c) Improve reporting of research using animals d) Improve the communication of the research findings to the broader scientific community e) Promote reproducible, transparent, accurate, comprehensive, concise, logically ordered, well written manuscripts
a
With regards to the hierarchy of scientific methods, which of the following lists the correct order from strongest to weakest? a) Meta-analyses & systematic reviews; randomised control trials; cohort studies; case control studies; cross-sectional studies; case reports b) Meta-analyses & systematic reviews; randomised control trials; cohort studies; crosssectional studies; case control studies; case reports c) Meta-analyses & systematic reviews; randomised control trials; cross-sectional studies; cohort studies; case control studies; case reports d) Meta-analyses & systematic reviews; randomised control trials; cohort studies; case control studies; case reports; cross-sectional studies e) Meta-analyses & systematic reviews; randomised control trials; case control studies; cohort studies; case reports; cross-sectional studies
a
What is quantitative research?
a formal, objective, systematic process in which numerical data are used to obtain information about the world.
Scatterplot
a graphed cluster of dots, each of which represents the values of two variables
box and whisker plot
a graphic way to display the median, quartiles, and extremes of a data set on a number line to show the distribution of the data. Box ends are 25th (1st quartile) and 75th percentiles (3rd quartile) and the median lies inside the box, whiskers show the smallest and greatest value
What is a forest plot?
a graphical display of estimated results from a number of scientific studies addressing the same question, along with the overall results
One-way ANOVA
a hypothesis test that includes both one nominal independent variable with more than two levels and a scale dependent variable
Two-way ANOVA
a hypothesis test that includes two nominal independent variables, regardless of their numbers of levels, and a scale dependent variable
What is meta-regression?
a meta-analysis that uses regression analysis to combine, compare, and synthesize research findings from multiple studies while adjusting for the effects of available covariates on a response variable; allows the effect of continous or categorical variables to be investigated
What is matching/stratification in research?
a method of sampling that involves dividing a population into smaller groups-called strata
What is a representative sample?
a sample that resembles the population
What is the declaration of Helsinki?
a set of ethical principles regarding human experimentation developed for the medical community by the World Medical Association (WMA)
What is a reporting guideline?
a simple, structured tool for health researchers to use while writing manuscripts. provides a minimum list of information needed to ensure a manuscript can be understood, replicated, used and included in a systematic review
Pearson's r
a statistic that measures the direction and strength of the linear relation between two variables that have been measured on an interval or ratio scale
What is regression analysis?
a statistical process for estimating the relationships among variables moving beyond correlations, to see what factors confound your results
What is a systematic review?
a structured, comprehensive synthesis of the research literature to determine the best research evidence available to address a healthcare question.
What is truncation?
a technique that broadens your search to include various word endings and spellings - i.e. cut something off · Eg: psychology* -- retrieves psychology, psychologist, psychological, etc.
What is the GRADE approach?
a transparent framework for developing and presenting summaries of evidence and provides a systematic approach for making clinical practice recommendations.; how to score quality of evidence
What are the advantages of quantitative interviews?
ability to collect in depth information, can clarify meaning, can be tailored to individuals
What is sensitivity in a systematic search?
ability to correctly identify relevant articles
Post hoc tests
additional significance tests conducted to determine which means are significantly different for a main effect
What is the Bonferroni correction?
adjustment made to p-value, accounts for problems of multiple data comparisons sets significance cut off at alpha/n where a - 0.05 and n is number of hypthothesis being tested
What are the functions of the National Committee for the Protection of Animals Used for Scientific Purposes?
advise HPRA and animal welfare bodies on acquisition, breeding, accommodation and care and shares best practices with other national committees in EU
What are randomization strategies with a high risk of bias?
alternate sequence data of birth/admission etc.
ANOVA stands for
analysis of variance
What are interventional studies?
animal studies and human studies (clinical trials), open studies, rcts etc.
Why would you use a non-parametric test?
area of study is better represented by the median you have a very small sample size in your study you have ranked data, ordinal data, or outliers which cannot be removed you have limits of detection
what font should you use for paper based questionnaires (i wish it was comic sans)
arial size > 12
What directive makes it mandatory for the HPRA to carry out an evaluation of every project application for animal research before issuing authorisation?
article 38 of directive 2016/63/EU requirement 31
How to make an effective questionnaire?
ask relevant questions get valid responses (reflect opinion of participants) get representative response
What can observational studies determine?
associations, not cause or effect
. Meta-analysis is a type of statistical analysis used in systematic reviews. What is the minimum number of studies needed to undertake meta-analysis? a) 1 b) 2 c) 3 d) 5 e) 10
b
A repeated measures study was conducted. A sugary drink was given to 10 mice and their behaviour was measured in an open field test, Later that week the same 10 mice were given a placebo and the researcher was interested in looking at the difference between their scores at the two stages, therefore a paired t-test was used to analyse the data. The pairedsample t-test is appropriate when: a) The variable of interest is binary b) We want to compare two numerical variables when each is measured on every individual in the sample c) We wish to test the null hypothesis that the mean of the differences between the pairs of observations in the sample is equal to zero. d) We wish to test the null hypothesis that the median of the differences between the pairs of observations in the sample is equal to zero. e) We wish to test the alternative hypothesis that the mean of the differences between the pairs of observations in the sample is equal to zero
b
A researcher is testing the effect of drug A on intestinal inflammation induced by DSS in rats. The researcher therefore is comparing of animals that received the drug A (DSS + drug A). If the researcher obtained a p value of 0.02, that would mean; a) To reject the null hypothesis and conclude there is no significant difference between mean inflammatory scores in the 2 groups of animals b) To reject the null hypothesis and conclude there is a significant difference between mean inflammatory scores in the 2 groups of animals c) To fail to reject the null hypothesis and conclude there is no significant difference between mean inflammatory scores in the 2 groups of animals d) To fail to reject the null hypothesis and conclude there is a significant difference between mean inflammatory scores in the 2 groups of animals e) To accept the alternative hypothesis and conclude there is no significant difference between the mean inflammatory scores in the 2 groups of animals
b
A researcher studied the relationship between the levels of tumour necrosis factor alpha and the histological damage in rats with colitis induced by DSS. The correlation coefficient of r = 0.929. This indicates: a) The two parameters are directly corellated and the link is weak as r is positive and close to 0. b) The two parameters are directly correlated and the link is strong as r is positive and close to 1 c) The two parameters are inversely correlated and the link is strong as r is negative and close to 1 d) The two parameters are inversely correlated and the link is weak as r is positive and close to 1 e) The two parameters do noy represent paired observations
b
A researcher wants to test the effect of drug A in intestinal inflammation induced by DSS in rats. He takes 2 independent groups of rats; 0 rats are given DSS and placebo ands 10 rats are given DSS and Drug A. The researcher was interested to see if drug A was able to reduce significantly the intestinal inflammatory score in treated rats compared to placebo. Select one of the following statements which you believe to be true. a) A paired-sample t-test should be performed on these data as the two groups are dependent b) A two-sample independent t-test should be performed on these data as the two groups are independent c) If the results were significant at the 5% level we could conclude that the null hypothesis cannot be rejected d) A Tukey's post hoc test should be performed after a two-sample t-test to know where the difference between both groups lies e) A one-way ANOVA should be performed on these data
b
As part of a randomised control trial participants are randomly allocated to either an intervention or control group. Which of the following is not considered an adequate method of allocation? a) Coin tossing b) Date of birth c) Drawing lots d) Random number generator e) Random number table
b
The I2 statistic is used in meta-analyses to assess which of the following? a) Effectiveness b) Heterogeneity c) Risk of bias d) Study quality e) Reporting
b
What is the name of PubMed's Controlled Vocabulary that can be used in a Search Strategy for a Literature Review? a) Advanced search b) MeSH c) Emtree d) Single Citation Matcher e) Title/Abstract
b
What non-parametric test is used to assess the difference between more than 2 independent samples? a) One-way NOVA b) Kruskal Wallis test c) Wilcoxon sign-rank test d) Pearson correlation e) Factorial ANOVA
b
Which of the following best describes reliability in relation to study findings? a) Extent to which the study findings are applicable to overall population of interest b) Extent to which the study findings are repeatable c) Extent to which the study findings report the research question posed d) Extent to which the study findings agree with the published literature e) Extent to which the study findings refute the published literature
b
Which of the following describes a complex intervention? a) One that is delivered by more than one healthcare professional b) One that has two or more interacting components c) One that always involves a patient d) One that has four interacting components e) One that is developed without the use of theory
b
Which of the following is a parametric test? a) Kruskal Wallis test b) Pearson correlation c) Spearman rank correlation d) Type 1 error e) Wilcoxon sign-rank test
b
Which of the following is not a characteristic of quantitative analytical studies? a) They are designed to reach causal inference about hypothesised relationships b) They may be used to identify trends in patterns of medicine use over a defined time period c) They can be aimed to gain a deeper understanding of explanatory factors of drug prescribing dispensing and consumption d) They could be carried out to analyse the positive or negative effects of medication therapy e) They require a design permitting an evaluation of the causal effect associations between exposure and outcome
b
Which of the following is not true regarding population data? a) Lack of uncertainty in results deriving from fluctuations in sample results and sample errors b) Most studies are based on research from one or more samples selected out of the total population c) One of the disadvantages is the cost d) Results provide an accurate summary of that population e) Utilise the entire population
b
How to prevent detection bias?
blinding outcome assessments
How to avoid performance bias
blinding participants and personnel
What is an example of continous numerical (quantitative data)?
blood pressure
Are cohort studies retrospective or prospective?
both
1. SPSS is an acronym of the following: a) Sexual Performance and SAD Syndrome b) Sexual preferences for the sixties and seventies c) Statistical package for the social sciences d) Statistical predictions for the social sciences e) None of the above
c
In an observational study of defibrillation in theatre, 23 surgeons and 25 anaesthetists were asked to manage simulated ventricular fibrillation. Overall, the anaesthetists fared better than surgeons at managing ventricular fibrillation, although because of the small sample this difference in proportions was not significant (7/25 vs 1/23. P=0.06). which of the following is correct? a) This lack of significance could be due to a type 1 error b) This lack of significance could be due to a type 2 error c) The chance of a type 2 error increases as the sample size decreases d) A larger sample size would have reduced the type 1 error rate e) This lack of significance could be due to a gamma error
c
MCQ: We are interested in investigating whether a novel drug is effective as a weight reducing agent in rats. Thirty obese rats are randomly allocated to receive either the drug or placebo. Each takes the relevant preparation once a day for two months, whilst eating a normal diet, and their weight is measured at the end of the period. Select one of the following statements which would provide an appropriate null hypothesis for the study. a) At the end of the two month period, the mean weight in the placebo group is greater than that of the drug group b) At the end of the two month period, the mean weight in the placebo group is less than that of the drug group c) At the end of the two month period, the mean weight in the placebo group is equal to that of the drug group d) The mean change in weight from baseline to two months is greater in the drug group that in the placebo group e) The mean change in weight from baseline to two months is greater in the placebo group that in the drug group
c
We are interested in investigating whether a novel drug is effective as a weight reducing agent in rats. Thirty obese rats are randomly allocated to receive either the drug or placebo. Each takes the relevant preparation once a day for two months, whilst eating a normal diet, and their weight is measured at the end of the period. Select one of the following statements which would provide an appropriate null hypothesis for the study. a) At the end of the two month period, the mean weight in the placebo group is greater than that of the drug group b) At the end of the two month period, the mean weight in the placebo group is less than that of the drug group c) At the end of the two month period, the mean weight in the placebo group is equal to that of the drug group d) The mean change in weight from baseline to two months is greater in the drug group that in the placebo group e) The mean change in weight from baseline to two months is greater in the placebo group that in the drug group
c
What is the name of a study in which data are collected by 'observing' what is happening and is conducted in real life settings without any experimental set up? a) Clinical trial b) Experiment c) Observational study d) Pilot study e) Hypothesis
c
Which of the following statements best describes the 'convenience sampling' approach in qualitative research? a) The researcher makes initial contact with a small group of people b) The researchers see heterogeneity in a sample and aim to identify a diverse range of participants c) Participants are selected because they are readily available to the researcher d) The researchers seek heterogeneity in a sample and aim to identify similar participants e) The researcher does not seek heterogeneity in a sample
c
. Which of the following is not a characteristic of case-control studies? a) If a significant difference in exposure history is found, then can infer that risk exposure has some association with disease b) They are mostly retrospective c) They identify patterns or trends in medicine use but do not allow for inference to be drawn about causal associations d) They work from effect to cause e) Usually a set of people with the disease (cases) are 'matched' to a ser of people without disease 'controls'
c?
Advantages of face to face interviews vs. telephone interveiws?
can act on non-verbal ques and collect longer responses
What are the weaknesses of quantitative research?
can't provide deeper insights into behavior or phenomena
What kind of data is sex (M/F)?
categorical, nominal data (unordered categories)
What kind of data would be a grade of breast cancer?
categorical, ordinal data
What are examples of questionnaire studies?
census, health related quality of life, any questionnaire is a scientifically validated instrument
Methods of Allocation Concealment
central allocation by computer or third party using sequentially numbered, sealead, opaque envelopes
What tests can you use for group comparisons of data?
chi-squared t-tests correlations mann-u whitney etc.
What are characteristics of a good question in questionnaire?
clarity, minimise ambiguity, ensure accuracy and honesty, and eliminate potential bias
What are different types of heterogeneity?
clinical, methodological, statistical
What is the difference between cohort studies and case control studies?
cohort study is concerned with frequency of disease in exposed and non-exposed individuals, the case-control study is concerned with the frequency and amount of exposure in subjects with a specific disease (cases) and people without the disease (controls).
What are the advantages of a focus group?
collect in depth info, clarify meaning, group setting can stimulate new ideas, generate ideas, understand customers' vocabulary, and reveal consumer needs and attitudes, mixed background of participants
What is Cronbach's alpha?
common way of computing correlation values among the questions on instruments, measures internal consistency
What is a box and whisker plot used for?
compare two groups of numerical data visually e.g. time it took for boys vs. girls to complete a race at school
T-test and ANOVA
compares differences in the means of 2 groups, when data is interval or ratio (analysis of variance) - compares the means of 2 or more groups
Repeated ANOVA
compares means across one or more variables that are based on repeated observations. A repeated measures ANOVA model can also include zero or more independent variables. Again, a repeated measures ANOVA has at least 1 dependent variable that has more than one observation. one group compared on 3 or more different measures
What is allocation concealment?
concealment of an experimental group assignment to participants and investigators delivering the intervention in an RCT limits selection bias done before the individuals are grouped into their level
What does the width of the horizontal line through diamond in a forest plot estimate?
confidence interval
what does the horizontal line on a forest plot represent?
confidence interval (CI)
What kind of data is height?
continous, quantitative (numerical)
if the diamond in a forest plot is to the left, what can you conclude?
control is better
How do you test for correlation between two sets of data?
correlation co-efficient; strength of linear association between two variables
what can you use to check internal consistency of a questionnaire?
cronbach alpha
. According to the International Council for Harmonistaion (ICH) guideline on Good Clinical Practice, in a clinical trial priority (over other interests) should be given to: a) The interests of science b) The interests of society c) The interests of society and science equally d) The interests of the trial subjects e) The interests of the trial subjects and society equally
d
. In a study, 30 rate are randomly assigned to one of three groups; control (n=10), treatment A (n=10), or treatment B (n=10). After treatment, the eman scroes for the three groups are compared, comparing these means is: a) Repeated measures ANOVA b) The correlation coefficient c) Factorial ANOVA d) One-Way ANOVA e) Two-Way ANOVA
d
How many times do you have to repeat an in vitro experiment to make the data valid? a) At least 30 times b) At least 20 times c) At least 10 times d) At least 3 times e) There is no need to repeat an in vitro experiment to make the data valid
d
In an experiment, a researcher wants to study of salted drink water could affect the blood pressure in mice. What is the 'dependent variable' in that particular experiment? a) The investigator b) The cages where animals are housed c) Salted drinking water d) Mice blood pressure e) The research being carried out
d
MCQ: You are presented with a dataset containing numerical variables 'number of medicines' and 'age in years', what would be the most appropriate way to present the data? a) Cross-tab b) Histogram c) Pie chart d) Scatterplot e) Venn diagram
d
Subjects from families whith genetic disorders were asked whether they had encountered problems when applying for life insurance. A sample from the general population was also asked the same question. All in the study group (723/2167) reported problems compared with only 5% in the general populations. This difference was significant at the 0.01% level. Which of the following statements you believe to be true? a) The suitable null hypothesis would be that subjects from families with genetic disorders in the population are more likely to experience problems when applying for life insurance than those from the general population b) The result is significant at the 0.01% level. This means that we are 0.01% certain that the null hypothesis is true. c) The difference between the two groups of families are likely to have occurred by chance d) We can reject the null hypothesis at the 5% level of significance e) We can reject the alternative hypothesis at the 5% level of significance
d
What is the name of an effect that occurs when an experimental group gets better simply because the are being given a pill and this leads them to expect to get better? a) The butterfly effect b) The domino effect c) The expectancy effect d) The placebo effect e) The treatment effect
d
When developing a questionnaire, which question styles are preferable? a) Only open questions b) Only closed questions c) Only Likert scale measurement questions d) A combination of closed questions and likert scale questions e) True or false questions
d
Which is the control variable in an experiment? a) It is what is varied during the experiment b) It is what will be measured c) It is what the investigator thinks will affect the independent variable d) It is the variable that an investigator keeps constant to prevent the possibility that other factors could affect the outcome e) It is the statistical analysis
d
Which of the following ANOVA tests compares means across one or more variables that are based on repeated observations? a) One-way ANOVA b) Two-way ANOVA c) Three-way ANOVA d) Repeated measures ANOVA e) T-test ANOVA
d
Which of the following is not a characteristic of inferential statistics? a) Help to assess the strength or magnitude of relationships (association) between independent (or causal) variables and dependent (effect) variables b) Help to infer (discover) properties or characteristic, or general pattern about a large group by first studying a smaller group, so those results can then be generalised back to a larger group c) An example would be smoking as independent variable and lung cancer as the dependent variable (possible outcome of smoking) d) Provide a summary of the data gathered in mathematically numerical form or visually e) All of the above are characteristics of inferential statistics
d
You are presented with the number of medicines takekn by 12 patients; 3, 4, 4, 4, 2, 1, 4, 8, 2, 4, 1, 4, 7. What is the mode of these data points? a) a)2 b) 3 c) 3.8 d) 4 e) 8
d
Why is data entry harder for paper vs. online surveys?
data entry required
What document is used for data extraction?
data extraction form
What is categorical nominal data?
data that has two or more categories, but there is no intrinsic ordering to the categories.
What is parametric data?
data that resembles a known hypothetical statistical distribution --> i.e. the normal distribution
What are the aims of a systematic review protocol?
describe the plans to be followed and minimise bias by stating a priori hypotheses and methods without advance knowledge of results
What is the purpose of discussion and conclusion section in research?
describes main findings and why work is important, summarises key findings, implications and future directions
Is a retrospective chart describing how frequently prescribers in a hospital ward use an antibiotic, descriptive or inferential?
descriptive
We use _____ statistics simply to describe what's going on in our data.
descriptive
What is an invitation letter for a focus group?
details of study, brief background, how long interview will last, contact details
what is the methodology of a search strategy?
determine clear and focused question describe articles that can answer that question decide on key concepts that address elements of that question decide which elements should be used for best results choose an appropriate database and interface to start with document search process identify appropriate index terms in thesaurus of that database identify synonyms add variations use appropriate syntax as per database optimise evaluate initial results check for errors translate to other databases retrieve
What does a t-test tell us?
difference in means of 2 groups
What is the Chi-square test?
differences in frequency between two categorical variables - checks whether a distribution is significantly different than an expected or theoretical distribution
How to address heterogeneity?
double check data is correct do not do a meta-analysis explore heterogeneity - subgroup analyses perform a randoms effects meta-analyses (incorporates heterogeneity, especially if it cannot be explained) reconsider the effect measure exclude studies (could perform a sensitivity analysis and identify studies but try not to introduce bias)
What is theming in qualitative data analysis?
drawing codes from one or more transcripts to present findings in a coherent and meaningful way
As part of a randomised controlled tria; investigating the effectiveness of a novel drug treatment for COVID-19, participants were not blinded to treatment allocation. Which of the following risk of bias is it? a) Attrition bias b) Detection bias c) Performance bias d) Reporting bias e) Selection bias
e
MCQ: Which of the following is not a quantitative descriptive study? a) Case-control study b) Case reports c) Cross-sectional study d) Longitudinal observational study e) All of the above are quantitative descriptive studies
e
Which of the following is an example of a discrete variable? a) Age (in years) b) Blood group (O, A, B, AB) c) Blood pressure d) Height e) Number of children
e
Which of the following is not characteristic of a good experiment? a) Simple b) Blinding c) Randomisation d) Replication e) All of the above are characteristic of a good experiment
e
An academic and a hospital consultant are collaborating on a research project that includes a survey of the consultant's patients. A final year undergraduate and a postgraduate undertake the academic survey to patients who wish to participate, when patients attend routine clinic appointments scheduled by a hospital administrator. Which of the following would be the most appropriate gatekeeper? a) The academic b) The consultant c) The postgraduate d) The undergraduate e) None of the above
e (answered b? but best not to have anyone involved in research process???)
How do you improve validity of a questionnaire?
e.g. giving the questionnaire to the same group of respondents at a later point in time and repeating the research. Then, comparing the responses at the two time points.
What is an odds ratio?
estimate of relative risk that is used in case-control studies. how much more likely it is that a person with a disease has been exposed to a risk factor than someone w/o the disease
what must be obtained before submission of project application to the HPRA for animal research?
ethical approval
What is internal validity in quality and risk of bias assessment?
extent to which design and conduct of a study are likely to have prevented bias
what is external validity in terms of quality and risk of bias assessment?
extent to which results provide a correct basis for generalizations to other circumstances
What is an independent variable?
factor in an experiment that a scientist purposely changes
What is a case-control study?
finds aetiology - observational study where 2 groups that have different outcomes are compared to find a causal factor e.g. non-exposure to benzene vs. exposure and how it relates to leukaemia
What is a non-parametric test?
for skewed distribution, looks at some other comparison between groups e.g. comparing the ranks of values instead of the values themselves
ICH GCP - before a trial is initiated?
foreseeable risks and inconveniences should be weighed againsts anticipated benefit for the individual trial subject and society
What is step 2 of the scientific method?
gather information
What is secondary research?
gathering information that already exists about the topic in question (library or internet research) getting information second hand that has already been collect by someone else e.g. systematic reviews, economic analysis
What is academic dissemination?
getting the findings of your research to the people who can make use of them, to maximise the benefit of the research without delay
What is an I2 statistic?
gives you an idea of the heterogeneity of the studies, i.e. how consistent they are
What are aims and why do we have objectives?
goals that the research wants to achieve, and having objective helps attain those aims (steps to take to achieve aim)
What types of analyses can you do in qualitative research?
grounded theory, framework method, thematic analysis or content analysis
What must you consider when performing a meta-analysis?
has >1 study estimated a treatment effect for the outcome of interest are study characteristics sufficiently similar what type of data was used what model will best suit to pool results
Risk of detection bias low/high/unclear; open-label trial design
high
Risk of incomplete outcome data bias low/high/unclear; a non-significant proportion of patients dropped out prior to the end of study in the treatment group (60%) compared with placebo group (20%)
high
Risk of reporting bias low/high/unclear; not all study's primary outcomes have been reported
high
What is an acceptable response rate?
higher better, 30 - 50%
What graphs can you use for numerical variables?
histograms box plots (box and whisker) scatterplots
What are the strengths of quantitative research?
hypotheses may be confirmed, results can be inferred for the population
What is coding in qualitative data analysis?
identification of topics, issues, similarity and differences revealed through participants narratives and interpreted by researcher
What is a descriptive observational study?
identifies patterns or trends
When can you carry out research on a vulnerable group?
if research is responsive to health needs or priorities of this group and cannot be carrried out in non-vulnerable group, and participants benefit from the results of research
What will a One way ANOVA measure?
if there is a difference between three or more independent groups; but we don't know where the difference lies - use post-hoc to determine which mean(s) are different
What does a 95% confidence interval measure?
imprecision around the sample result
What are the principles of ICH GCP in terms of ethical principles?
in accordance with declaration of Helsinki
Selection Bias
in an experiment, unintended differences between the participants in different groups
What are double negative questions?
includes two negative words, potentially confusing or misleading the participant completely e.g. what is not unlikely my favourite flavour of monster
What are issues in secondary research?
incomplete database, subjective analysis of data, poor analysis in which variety of techniques are used
How do you minimise chance effects (random error)?
increase no. of participants
Why would you perform a meta-analysis?
increase power increase precision settle controversies
We use _____ statistics to try to infer from the sample data what the population might think. Or, we use ____ statistics to make judgments of the probability that an observed difference between groups is a dependable one or one that might have happened by chance in this study. Thus, we use ______ statistics to make inferences from our data to more general conditions
inferential
Outcome Data
information gathered after a change has occurred to examine the impact or results
What are randomisation strategies with an unclear risk of bias?
insufficient information provided to permit judgement e.g. just saying 'patient were randomly allocated'
What does ITT stand for?
intention to treat
What is PROSPERO
international database of prospectively registered systematic reviews in health and social care.
if diamond in forest plot is to the left, what can you conclude?
intervention is better
What is the interview effect/response bias?
interviewer's expectations, beliefs, and prejudices as they influence the interview process and the interpretation of the data it provides.
What is ethnographic research?
investigators immerse themselves in the situation being studied to observe and describe naturally occurring phenomena.
incomplete outcome data
is there missing data due to attrition or exclusions from analysis
What is a fixed-effect model?
it is a model used to pool different studies that have the same question and have results that only differ by chance (i.e. no statistical heterogeneity), each study weighted by its sample size (more weight on bigger studies)
In a forest plot, what is the size of the box proportional to?
its weight
What are the three principles of intention to treat analysis
keep participants in intervention groups to which they were randomised, regardless of the intevention they actually received measure outcome data on all participants include all randomised participants in the analysis
What do whiskers in a forest plot represent?
length of confidence intervals
What is a cross-sectional study?
looking at gorup of different people at one moment in time, analyzes data from that group
Do highly sensitive search strategies have high or low precision?
low
Risk of bias low/high/unclear; study was performed using a double blinded design, both participants and assessors who did blood pressure testing did not know which group participants belonged to
low
Risk of detection bias low/high/unclear; study used double blinded design and investigators who assessed treatment outcomes were blind to which group the partipants belonged to
low
Risk of incomplete outcome data bias low/high/unclear; ten of the recruited partipants dropped out of the study, they were equally divided across control and intervention arm
low
Risk of reporting bias low/high/unclear; study protocol available and study's pre-specified outcomes have been reported as per protocol
low
What is the risk of bias (low/high/unclear) with this allocation sequence; groups generated by statistician using a random number generator, assesors not involved in this aspect and details of allocation were kept in a locked filing cabinet
low
If a study has low power, what are the chances of the study finding a significant result when there actually is a significant result present, (i.e., type II error)?
low chance
What score does an observational study get with the GRADE approach?
low quality rating
What is an inductive reasoning process?
makes specific observations and draws general conclusions
what is the aim of the ARRIVE guidelines?
maximizes published information on animal studies and minimizes unnecessary studies, guidance on how to conduct a trial.
Issues with non-systematic reviews
may not answer clear research questions do not attempt to locate all information do not describe inclusion criteria don't critically appraise evidence potential for summary to be biased may not be reproducible
For normally skewed data, with a bell shape, what is the measure of central tendency?
mean
How would you measure the effect measure of continous data?
mean difference (MD)
What are the measures of central tendency?
mean, median, mode
What does the M in SMART goals stand for?
measurable- goals should be measurable so you have tangible evidence that you accomplished your goal. usually the entire goal statement is a measure for the project, but there are usually short term or smaller measurements built into the goal
What is trustworthiness in qualitative research?
mechanism to establish confidence in truth of findings - coding of info can be done by 2 researchers
for a skewed distribution, what measure of center tendency should be used for ordinal data?
median
What is an experimental study?
method of determining cause and effect by manipulating certain variables and observing the effect on some behavior
What is the framework method in qualitative research?
method used to organize and structure research data in order to refine the researcher's focus to identify salient themes
What is inferential statistics?
methods for drawing conclusions about a population by examining a *sample* of the population. ♦ Drawing conclusions from a sample of a population ♦ Measuring the reliability of conclusions drawn from samples.
What is descriptive statistics?
methods of organizing, summarizing, and presenting numerical data in an informative way
What is an intentional error?
mistake or violation - lack of expertises/failure or expertise or delibrate rule break
for a skewed distribution, what measure of central tendency should be used for nominal data?
mode
Strong, moderate or weak correlation, r = - 0.6
moderate
Strong, moderate or weak correlation, r = 0.6
moderate
how to calculate the mode?
most frequently occurring score/value within a data set. Sometime there are 2 modes (bi-modal) or no mode if all the scores are different.
ICH GCP - the rights, safety and wellbeing of trial subjects
most important considerations and should prevail over interests of science and society
Are case control studies prospective or retrospective?
mostly retrospective
What kind of setting does qualitative research take place in?
naturalistic/ethonographic research
Do you always have to use statistics (meta-analysis) in a systematic review?
no
are allocation concealment and blinding the same?
no
If correlation coefficient is 0
no association between the two variables
Is ethical approval a legal requirement for HPRA to approve a animal research project and provide authorisation?
no but speeds up the efficiency
What would it mean if the diamond (overall result of meta-analyses) crosses the middle line (line of no effect)
no statistically significant difference between the control and intervention arm
Is the National Committee for Protection of Animals Used for Scientific Purposes part of the HPRA?
no, independent body not subject to any direction/influence, provides information to HPRA and ensures best practice in Ireland, shares best practice with other committees
How is sensitivity calculated?
no. of relevant articles identified by search / total no. of relevant articles in existence
Difference between dunnet and tukey?
not all pairwise comparisons are made with Dunnet, only compare groups with 1 control group
What are the challenges of ITT?
not always possible to collect or include data on all participants due to attrition (dropouts/loss to follow up) also doesn't have a clear definition in literature e.g. authors may describe ITT even if some outcome data missing
What is a null hypothesis?
null means zero. so a null hypothesis states that two variables are not related; i.e. orginal hypothesis is incorrect
What does the unbroken vertical line through centre of a forest plot represent?
null value
How do you calculate precision of search?
number of total relevant articles retrieved / total number of articles
what are behavioural change techniques?
observable, replicable and irreducible component of an intervention designed to alter or redirect causal processes that regulate behaviour
What are the STROBE guidance used for?
observational studies
Regression co-efficient
obtained from a meta-regression describes how the outcome variable (intervention) changes with a unit increase in the explanatory variable (potential effect modifier)
Reporting bias
occurs when a source has the required knowledge but we question his or her willingness to convey it accurately - selective outcome reporting
Attrition Bias
occurs when participants drop out of a long-term experiment or study
How would you measure the effect measure of dichotomous data?
odds ratio (OR)
In an ANOVA, rejection of H0 tells you that there is a high probability that at least
one difference exists somewhere, not where that difference is
Benefits of Systematic Reviews
organise collection of evidence appraise quality minimise bias compare and contrast studies combine findings - increase statistical power improve cost effective interventions design better studies
What should a data extraction form include?
outcome data, assessment of bias, additional data e.g. demographics, intervention types
What is a schedule or topic guide?
outline of key issues and areas of questioning used to guide a qualitative interview or group discussion to explore the topic of interest - can be the aim of the study
What is the purpose of an introduction in research?
outline topic (what is known/not known, conflicting views, supportive evidence) and why it is important to conduct research
What does the diamond in a forest plot represent?
overall summary estimate; centre represents the pooled point estimate
Consequences of low power
overestimation of effect sizes and lack of reproducibility of results, unethical research and wasteful
Difference between paired and unpaired t-tests?
paired - at two different times (same subject) unpaired - two different subjects/experimental units
What test to use; numerical data, two groups, paired
paired t-test
What test would you use; measure of hair length of people before or after lockdown - subjects hair measured at start and end.
paired-t-test (same subjects, data not independent but from same subjects),
What are the types of distribution methods for surveys?
paper electronic face to face telephone
What is a randomized control trial?
participants are recruited and are randomly assigned to a treatment/exposure and comparing outcomes between groups over time. - by randomizing, you are eliminating all other confounds because the only difference between groups is the intervention they receive
Randomisation ensures what?
participants have an equal chance of being allocated to either intervention/control arm
What is the difference between population and sample data?
pd: utilises the entire population, sample based on research from one or more samples
If you have parametric data, what test should you use for correlation?
pearson 'r'
What does PPT stand for?
per protocol analysis - analysis of subset of participants who complied with the protocol sufficiently to ensure their data would be likely to exhibit the effect of treatment
If correlation co-efficient is -1, what is the strength of linear association?
perfect negative correlation - no correlation between data sets
If correlation co-efficient is +1, what is the strength of linear association?
perfect positive correlation, the two data sets are highly correlated
What graphics can be used for categorical variables?
pie charts bar charts venn and euler
What is a random effects model?
pooling together studies that asked different questions but are similar enough to compare the data from; i.e. treatment effect for individual studies vary around an overall central effect, allows for heterogeneity
Where would you take averages or standard deviations from?
population parameters
What is primary research?
practice of gathering the original data yourself by means of observations, surveys and experiments.
Tukey post hoc test
preferred test - greatest power, pairwise compairson of means
What is power in hypothesis testing?
probability of rejecting the null hypothesis when it is false; most common reason for type 2 errors is that the study is too small probability of making a correct decision (to reject the null hypothesis) when the null hypothesis is false. 1-type2 error i.e. 1-β
What is 'power' of a study?
probability that the study will find a statistically significant difference when one is truly there. Increasing the sample size increases power
How to avoid selection bias?
proper allocation sequence generation allocation sequence concealment
What is risk in terms of dichotomous data?
proportion of participants with outcome of interest e.g. outcome is 40/100, risk is 0.4
what is precision in a search?
proportion of relevant articles identified
What documents are required for an interview/focus group study?
protocol schedule/topic guide invitation letter information sheet consent form
What is the purpose of results section in research?
provide evidence for the research objectives
What is a power analysis?
provides an estimate of the minimum number of respondents needed to provide the best chance to discover if something does or does not exist
What should an information sheet contain for a qualitative study?
purpose, why chosen, if they have to take part and what will happen, confidentiality methods, who is organising/funding, risks and benefits
ICH GCP - who is responsible for medical care of participants?
qualified physician, or dentist
What kind of data is number of children in a study?
quantitative, discrete
What are open-ended questions?
questions that allow respondents to answer however they want
Weak correlation, what values for r?
r < 0.5 and r > -0.5
Strong Correlation = r> what and r < what
r > 0.8 and r < -0.8
What method should you use in meta-analyses if studies contain lots of heterogeneity?
random effects
What are randomisation strategies with low risk of bias?
random number generator random number table coin tossing
Which is the most reliable method of creating homogenous treatment groups without involving any potentianal biases or judgement?
randomisation
How do you minimize systematic biases (systematic error)?
randomisation, blinding, use analysis based on allocated treatement (intention to treat), include all relevant evidence
characteristics of strong allocation sequence
randomly generated and concealed
What is a confidence interval?
range of values in which a specified probability of the means of repeated samples would be expected to fall
What is relative risk?
ratio of risks in two groups
Why is it important to consider grey literature for EBP?
reduce publication bias, increase reviews' comprehensiveness and timeliness and foster a balanced picture of available evidence
What is selection bias?
refers to bias in how people are chosen to participate
Attrition bias
refers to the systematic loss or withdrawal of participants from follow-up
What is study heterogeneity?
refers to the variation in study outcomes between studies = if studies are too different it does not make sense to pool the data together i.e. comparing apples to oranges does not make sense (good for meta-analysis)
If I2 is 65%, would you reject or accept the fixed effects assumption?
reject
p-value < chosen significance, reject or accept null hypothesis?
reject
if p < 0.05, what does this mean?
reject null hypothesis and conclude statistical significance
type 1 error
rejecting the null hypothesis when it is true (false positive)
What is a narrative synthesis?
reliance on words and text to summarise the findings of the systematic review synthesis
What is reliability/reproducibility?
repeatable findings - the consistency with which the same event is repeatedly measured. scores are consistent across repeated testing.
What are the 3 R's?
replacement, reduction, refinement
What are cohort studies?
research looking at all of one group with shared characteristics and examining them
How should a questionnaire study be designed - key sections?
research questions -> aims and objectives -> study design (protocol, ethical approval, info leaflet, consent), questionnaire design, pilot -> questionnaire refinement, questionnaire distribution, analysis and disemmination
.What is qualitative participant observation?
researcher becomes a participant in culture/context, gains access
What is qualitative direct observation?
researcher unobtrusive so not to bias observation, can use technology and more detached
What are longitudinal studies?
researchers observe the same individuals (same variable) on many occasions over a long period of time
what are categorical questions?
response fits into one of several categories e.g. info relating to age
What is qualitative indirect observation?
results of an interaction, process or behaviour are observed
What are the functions of an animal research ethics commitee?
reviewing all stages of work to ensure it is adequate and justified, review ethos of establishment, ethical approval by assessing benefit vs. harm and ensuring 3Rs are fulfilled
if relative risk is greater than 1
risk in exposed greater than unexposed (positive outcome, intervention was not successful)
How do you calculate relative risk?
risk in treatment group/risk in control group
How do you achieve data saturation?
sample size big enough that no new themes are emerging from participants
What kind of bias is sample data subject to (compared to population data)?
sampling and selection bias
what are close-ended questions?
select from list of options - optioned choice/survey, limits the answer possibilities
What are the types of bias?
selection assignment assessment interviewer respondent recall information publication
What does matching/stratification prevent?
selection bias and confounding factors
What is an unintentional error?
slip or lapse - failure of attention or memory
if you have non-parametric data, what test should you use for correlation?
spearman rho
What does the S in SMART goals stand for?
specific- goals should be simplistically written and clearly defined. what are you going to do. the what, how, and why.
What are measures of variation in descriptive statistics?
standard deviation, range, or interquartile range
What is a meta-analyses?
statistical combination of results from at least 2 studies
What is a meta-analysis?
statistically pools the results from previous studies into a single quantitative analysis that provides the highest level of evidence for an intervention's efficacy - more precise effect of results
Strong, moderate or weak correlation, r = -0.9
strong
Strong, moderate or weak correlation, r = 0.9
strong
For inferential statistics a sample must be...
structured and consituted - randomised and representative and of sufficient size
If I2 is > 50%, what does this mean?
studies are inconsistent due to a reason other than chance.
What are the essential 10 in ARRIVE guidelines?
study design, sample size, inclusion and exclusion criteria, randomisation, blinding, outcome measures, statistical methods, experimental animals, experimental procedures and results
What is an analytical observational study?
study to find causal inference about hypothesised relationships, getting deeper understanding of explanatory factors
What is a parallel group design in RCT?
subjects are randomized to one or more study arms and each study arm will be allocated a different intervention. After randomization each participant will stay in their assigned treatment arm for the duration of the study
Does I2 > 50% represent substantial or insubstantial hetereogeneity?
substantial
Reporting Bias
systematic differences between reported and unreported findings
What is bias?
systematic error or deviation in results or inferences from the truth
Difference between systematic review and a scoping review?
systematic: uncover the international evidence, confirm current practice/variation/new practice, inform areas for future research and investigates conflicting results, produce statements to guide decision making scoping: identifies types of available evidence, clarifies key concepts/defintions, examines how research is conducted, identified key characteristics or factors related to a concept, precursor to systematic review main difference: scoping review will have a broader "scope" than traditional systematic reviews with correspondingly more expansive inclusion criteria
Look at differences between two groups on a variable e.g. gender/age
t-test
What is evidence based medicine?
the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients
What does statistical heterogeneity describe?
the degree of variation in effect estimates from a set of studies (measured by I2
What is validity in terms of risk of bias and quality assessment?
the degree to which a result is likely to be true and free of bias
What is validity?
the degree to which a test measures what it says it measures. e.g. the peabody test said it measured intelligence. however, it only tests receptive vocabulary, so that's not valid.
Interquartile Range (IQR)
the difference between the first and third quartiles
What is the mean difference?
the difference between the two means in the problem
What is a 95% confidence interval?
the interval in which we are 95% confident that the true value or our measurement can be found
standard error
the name for the standard deviation of a distribution of means
What is response rate?
the percentage of respondents who complete the questionnaire among those who are contacted/selected
What is the placebo effect?
the phenomenon in which the expectations of the participants in a study can influence their behavior
What is health research?
the process of systematically investigating a single well-defined aspect of physical, mental, or social well-being
standard deviation
the square root of the variance and provides a measure of the standard, or average, distance from the mean
What does inferential stats assess?
the strength or magnitude of relationships/associations between independent (or causal) variables
What is a monotonic relationship?
the value increases (or decreases) on one variable as the value of cases increases (or decreases) on the other variable.
Three-way ANOVA
three independent variables are in the ANOVA analysis
How many mean scores would you need for an ANOVA test?
three or more
What is a response burden?
time and effort required of participant to complete questionnaire, related to length of questionnaire and complexity of questions
Disadvantages of face to face vs. telephone qualitative interviews?
time consuming, expensive (researcher/participant time and travel, more accesible to other groups
What are the disadvantages of quantitative interviews?
time, smaller samples, expensive, potential for bias (interviewer effect)
What does the T in SMART stand for?
time-bound
What is an ANOVA test?
to compare mean scores of three or more groups within one study
When would you use qualitative over quanitative?
to explore areas not amenable to numerics, such as complex behaviours, attitudes and interactions
True or False: changes in protocols should be recorded and open to audit
true
True or false: box plots are more appropriate to compare and visualise three or more samples of numerical data?
true
What test would you use; comparing mean number of meds taken by patients in Dublin vs. in Kerry (two independent groups)
two-sample t test, want p < 0.05
What does a measurement scale provide?
type of ordinal data and typically has assessment questions, which is scored then summed to give a total score
For a fixed model what is the pooled result viewed as?
typical intervention effect
Risk of detection bias low/high/unclear; double blind randomised design no other information given
unclear
Risk of reporting bias low/high/unclear; study outcomes not clearly stated in methods section
unclear
What is risk of bias with this allocation sequence (low/high/unclear); patients were randomly allocated
unclear
Risk of bias low/high/unclear; double blinded randomised controlled design
unclear not enough information to permit judgement
if relative risk < 1
undesirable outcomes, risk in the exposed is less than risk in the non-exposed (negative association, possibly protective i.e. the intervention was successful)
What test would you use: does chocolate make mice happier - one group given chocolate + happiness mean scored, one not given choclate (control)?
unpaired t-test (two independent groups)
When should you use an apostrophe (put these in bc Shane is afraid of punctuation lololol)
use of ownership, it is/has = it's
What is the standardized mean difference?
used as a summary statistic in meta-analysis when the studies all assess the same outcome but measure it in a variety of ways (e.g. two studies might use two different scales to record quality of life)
What are the characteristics of good data in a questionnaire?
valid, reliable, unbiased, discriminating (having or showing refined taste or good judgement.)
what is a null value?
value that can be assigned to variables of any reference type and indicates that a variable refers to no object
What score does a case series or case report get with the GRADE approach?
very low quality rating
How can heterogeneity be informally from a forest plot?
visual inspection
Strong, moderate or weak correlation, r = - 0.4
weak
Strong, moderate or weak correlation, r = 0.3
weak
what must you consider to check for attrition/incomplete data as source bias?
were participants who were randomised to the intervention in the control group or intervention group what was extent of dropout/withdrawal and losses to follow up in intervention and control group.
When is Tukey's HSD most appropriate?
when ANOVA rejects null hypothesis and sample sizes are equal
When do you need to do a tukey?
when ANOVA reveals significant results - significant F value
what is a cross over study?
when the patient uses her or himself as a control (half way through the study switch patients from treatment to placebo)
What is binary data?
when there are only two categories (not 0 or 1) but yes/no or male/female
Dunnett's Test
when you only compare groups to a control value
What factors to consider when assessing for detection bias?
who is assessing outcome and was there a risk of bias in terms of the assessment (subjective or objective)
What should you consider to assess blinding as a score of bias?
who was blinded, was there a risk of bias in the actual outcome assessment
ICH GCP - Clinical trials should be conducted in accordance;
with ethical principles which have origins in declaration of Helsinki
What unit of analysis is used in qualitative research?
words, narratives
Can non-parametric tests be used for sample sizes <30?
yes
Is an animal welfare committee a statutory requirement?
yes must have one for every breeder, supplier and user
what are dichotomous questions?
yes or no questions
What symbol represents a type I error?
α
What symbol represents a type II error?
β
How do you calculate the variability of data in inferential statistics?
σ^2 (σ = type I error)
what are scoping reviews?
A broad review gathering studies on a single topic and providing some categorizing, without assessing methods
Who makes up an animal research ethics committee i.e. minimal expertise?
A designated veterinarian (DV) or expert in charge of advisory duties in relation to well being and treatment of animals, animal care and welfare officers, representative of research community/with animal research experience, public interest representative 'lay person', a statitiscian
Spearman Rho
A method for computing correlation, used primarily when sample sizes are small or the variables are ordinal in nature.
Kruskal-Wallis test
A nonparametric statistical test used to compare three or more unpaired (independent) samples where the outcome is either ordinal or continuous with a skewed distribution, basically a one way ANOVA for non-parametric data
Wilcoxon Signed Rank Test
A nonparametric statistical test used to compare two paired (dependent) samples where the outcome of interest is ordinal or continuous with a skewed distribution.
sampling bias (selection bias)
A problem that occurs when a sample is not representative of the population from which it is drawn.
What is grounded theory?
A qualitative research method based on the symbolic interactionist perspective of human behavior. Grounded theory research combines inductive and deductive research methods. With the use of inductive processes, theory emerges from the data. Deduction is then used to test theory empirically.
what is an abstract?
A short summary of what the article is about (250-600 words)
Paired t-test
A test designed to determine the statistical difference between two groups' means where the participants in each group are either the same or matched pairs.
2 sample t test
A test of significance for 2 means. Requirements: -Simple random samples were performed. -Samples are independent. -Both populations are normally distributed.
What is a parametric test?
A test that assumes the populations drawn from are normally distributed
Fisher Exact Probability Test
A test used to compare frequencies when samples are small and expected frequencies are less than six in each cell.
What is a hypothesis?
A testable, tentative answer to a question. Based on observations and knowledge.
What are the main guidelines for reporting in vivo animal experiments?
ARRIVE
What guideline must you follow for animal research studies?
ARRRIVE
What is an observational study?
An observational study measures the value of the response variable without attempting to influence the value of either the response or explanatory variables.
What is intention to treat analysis?
Analysis by original treatment assignment, not by actual treatment received "Once randomized, always analyzed" (i.e., by original treatment assignment) Secondary analysis based on "treatment received"
What is a subgroup analysis?
Analysis of subsets of randomized patients
What is step 6 of the scientific method?
Analyze data
What is per protocol analysis?
Analyzing outcomes only for those who received the full treatment as assigned.
What does ARRIVE stand for?
Animal Research: Reporting In Vivo Experiments
Who is responsible for evaluating ethical use of animals in research?
Animal research ethics committee
What three elements make up a risk=benefit analysis by an animal research ethics commitee?
Assessment of potential benefits, assessment of potential benefits being realised (i.e. is it realistic to be achieved), assessment of any harm or suffering
What does the A in SMART stand for?
Attainable- set goals that are within your physical limits.
What is deductive reasoning?
Beginning with generalized principle and moving to specifics
Performance bias
Bias that occurs due to systematic differences in care between treatment groups or in exposure to factors other than the intervention being studied e.g. researcher favors intervention group over control
What is controlled vocabulary?
Certain language or a set of terms that a search engine uses e.g. Pubmed uses MeSh terms
What is COREQ?
Consolidated Criteria for Reporting Qualitative Studies = covers the reporting of studies using interviews and focus groups
What is ordinal data?
Data that can be placed in a sequence (e.g. best, worst, okayish; oldest - youngest).
descriptive data
Data that describes an observation using words e.g. details of participants
What is continuous data?
Data that is numerical and can take any value in specified range e.g. range of age, range of height
For non-animal research what legislation is involved?
Depends on study design - clinical trials legislation, data protection and health research legislations, guidelines relating to ICH good clinical practice/equator network and local policies
what does DV stand for in terms of an animal research ethics committee ?
Designated Veterinarian
Mann-Whitney U test
Determines whether two independent (uncorrelated) means differ significantly when data are nonparametric i.e. distribution is skewed
What EU legislation controls animal research ethics?
Directive 2016/63/EU
range
Distance between highest and lowest scores in a set of data.
What kind of post hoc would you use; if you were a pussy
Dooley test
How to eliminate variability?
Double Blind experiments and control groups and placebos. Replication and randomisation.
How would you check if drop-out rate doesn't pose the risk of incomplete data bias?
Drop-out rate is even between control and intervention arm
What kind of post-hoc test would you use; you have three treatment groups and a negative group compared to positive control?
Dunnett
Where could you find information about other models to use instead of animal research?
EURL ECVAM
What does EQUATOR stand for?
Enhancing the QUAlity and Transparency Of health Research
What are the steps of CONSORT?
Enrollment, allocation, follow-up, analysis
What is refinement in animal research?
Ensure quality in breeding, accommodation and care and preventing pain, suffering, distress or lasting harm to animals.n
What does EURL ECVAM stand for?
European Union Reference Laboratory for Alternate to Animal testing - European Centre for validation of alternative methods
What are the HPRA primary functions in animal research ethics?
Euthorise establishments, authorise projects, ensure compliance with EU directive, inspection, monitor and record keeping, report to EU commission, liaise with member states, ensure the 3 RS are followed
What is a factorial design?
Experimental studies with two or more independent variables - In a 2 × 2 factorial design with placebo, patients are randomized into four groups: (i) to treatment A plus placebo; (ii) treatment B plus placebo; (iii) both treatments A and B; or (iv) neither of them, placebo only.
What is step 3 of the scientific method?
Form a hypothesis
ICH GCP - what kind of consent must be given before participants can be in a trial?
Freely informed
ICH GCP - what guidance should be followed for manufacture, handling and storage of investigational products?
Good Manufacturing Practice (GMP)
What does GRADE stand for?
Grading of Recommendations Assessment, Development and Evaluation
What are vulnerable groups?
Groups with an increased likelihood of being wronged or of incurring additional harm -Children -Mentally or emotionally ill -Physically disabled -Terminally Ill -Prisoners - pregnant women
What is the shorthand for a null hypothesis?
H0
Who enforces animal research ethics in Ireland?
HPRA
Who gives project authorisation in animal research?
HPRA
Who is responsible for monitoring animal research establishments?
HPRA
What are the general principles of the Declaration of Helsinki?
Health of px is 1st priority, duty to promote and safeguard health and wellbeing and rights of px reseerch only conducted by individuals with appropriate ethics, scientific training and qualifications
What is preferred in meta-analysis: heterogeneity or homogeneity?
Heterogeneity
Risk of bias low/high/unclear; open-label trial was conducted
High
what is the risk of bias (low/high/unclear) with this allocation sequence; patients were randomly allocated by participating physicians based on day of week they attended practice
High
What score does an RCT get using the grade approach?
High quality rating
what are the levels of evidence?
I- High Quality RCT's, Systematic Reviews of RTC's, Meta-Analysis II- Small-Scale RCT's, non-randomized studies with a control group III- Non-randomized studies with one group (no control group) IV- Descriptive Studies (single subject designs, case studies & surveys) V- Expert Opinion, Literature Reviews
How can heterogeneity be measured formally from a forest plot?
I2 statistic
What are the ICH GCP guidelines?
ICH good clinical practice guidelines are a set of guidelines that have principles that must be fulfilled
Detection bias
If outcomes assessed are different between the intervention and control group
What is the difference between observational and experimental research?
In experiments, the researcher will undertake some experiment and not just make observations. In observational study, the researcher simply makes an observation and arrives at a conclusion.
ICH GCP - who ensures a trial protocol is approved and favourable?
Institutional Review Board and Independent Ethics Committee
What does ICH stand for
International Council for Harmonisation of technical requirments for pharmaceuticals in human use
What is a sensitivity analysis?
Involves testing how changes in one variable affect an outcome when all other variables are kept the same. By repeating the process for each of the variables one at a time you can assess which has the greatest impact e.g. omit study with high risk of bias
What criteria must be met for project authorisation in animal research?
Justified from scientific/educational or legal point of view, the purpose justifies the use of animals, designed to enable procedures to be carried out in most humane and environmentally sensitive manner.
Advantages of questionnaires vs. interviews
Lower costs Possibility of anonymity, greater privacy Lack of interviewer bias can be piloted and validated useful for assessing variables like attitudes, beliefs and quality of life
What is MEDLINE?
Medical Literature, Analysis and Retrieval System Online - Best known database from the NLM. It includes bibliographic listings for publications in the areas of medicine, dentistry, nursing, pharmacy, allied health, and veterinary medicine. Used to locate articles on HIM issues as well as articles on medical topics necessary to carry out quality improvement and medical research activities.
What is experimental design?
Method of organising experiment to ensure the right data and enough of it is available to answer the questions of interest clearly and efficiently as possible. Reduces sources of variability.
Who advises the HPRA on matters related to animal welfare and advises animal welfare bodies in institutions?
National Committee for Protection of Animals Used for Scientific Purposes
Can you perform a post-hoc test if the H0 is NOT rejected in ANOVA?
No
Is data limited to outcomes and results of studies included in a systematic review?
No
what is non-parametric data?
Nominal Ordinal Continuous but not normal distributed
What document is published for animal research in terms of authorised projects?
Non technical summary reports
When would there be a high risk of reporting bias?
Non-significant findings are not reported, as per protocol.
What does reduction mean?
Number of animals used at a minimum without compromising the objectives of the project?
What are the two main types of quantitative studies?
Observational or Experimental (Interventional)
Assessment of difference between two independent samples, what is used for a parametric dataset and non-parametric?
P - Independent Sample T-test NP - Mann-Whitney U Test
Assessment of difference between more than 2 independent samples, what is used for parametric and non-parametric?
P - One-way ANOVA NP - Kruskal-Wallis Test
What is used for assessment of association between relationship of two numerical datasets, for parametric and non-parametric?
P - Pearson Correlation NP - Spearman Rank Correlation
Assessment of difference between two paired data set, what is used for parametric and what is used for non-parametric data
P - paired sample t test NP - Wilcoxon Sign Rank Test
what guideline for a systematic review
PRISMA
Difference between parametric and non-parametric data?
Parametric statistics are based on assumptions about the distribution of population from which the sample was taken. Nonparametric statistics are not based on assumptions, that is, the data can be collected from a sample that does not follow a specific distribution
What is purposive sampling?
Participant chosen because they belong to a specific group e.g. they have ADHD/are a twin
What is blinding?
Participants and/or caregivers unaware of treatment given, so changes in behaviour should be similar and same care and diagnostic workup will be similar
What test to use; want to determine the association between calcium intake (mg/day) and knowledge about calcium (out of 50)
Pearson's correlation (r bigger than 0.8 and r is less than -0.8)
What is study power?
Power is the probability that a test will reject the null hypothesis correctly (calculated *power = 1 - beta*). (ex: 80% power will detect a true difference 8 out of 10 times, and possibly miss a true difference 2 out of 10 times). Increasing study power requires a larger study population
What does PRISMA stand for?
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
What are probing questions?
Probing questions follow up on what has just been discussed in order for the interviewer to learn more
Which should be done first quantitative or qualitative to complement the second research?
Qualitative
What is the difference in role of theory between qualitative and quantitaive research?
Qualitative: generates theories for futher study Quantitative: tests theories to confirm/refute
What is the difference in reasoning process between qualitative and quantitaive research?
Qualitative: inductive (drawing conclusions by going from the specific to the general) Quantitative: deductive (drawing conclusions by going from general to the specific)
What is the difference in methods between qualitative and quantitaive research?
Qualitative: interviews, focus groups, observation Quantitative: experiments, RCTS, surveys
What is the difference of researcher stance between qualitative and quantitaive research?
Qualitative: subjective Quantitative: Objective, unbiased, natural
What is the difference in strength between qualitative and quantitaive research?
Qualitative: theorectical Quantitative: statistical
What is the difference in sampling method between qualitative and quantitaive research?
Qualitative: theorectical Quantitative: statistical
What is randomisation?
Randomisation can refer to the random allocation of participants to conditions to help control for variation due to participants. / Secondly, it can refer to randomising the order in which participants take part in conditions (thus achieving a similar effect to counterbalancing). / Thirdly, it can refer to randomising the order of stimulus materials for each participant e.g. a word list in a memory experiment might be given on a different order to each participant.
What does the R in SMART stand for?
Realistic- manage your expectations when you set goals.
Why might PPT analysis be subject to bias?
Reasons for non-compliance to the protocol may be related to the treatment
What can be added once the ARRIVE essential 10 are fulfilled?
Recommended Set
What is step 5 of the scientific method?
Record and organise data
What are Boolean operators?
Refined searches with operators such as AND and OR, NOT
What is grey literature?
Research that has not been commercially published.
What items should be assessed in a risk of bias assessment?
Sequence generation (randomisation) Allocation sequence concealment Blinding of participants/personnel Blinding of outcome assessors Incomplete outcome data Selective outcome reporting
What test to use; want to find association between how helpful information was (4-point likert scale - helpful to not helpful) and the degree of adherence to statins (indirectly measured by on-time prescription refill monthly rates %)
Spearman Rho
Difference between fixed effects and random effects model?
Studies are weighted differently - why? fixed effect analysis we assume that all the included studies share a common effect size
What is convenience sampling?
Subjects are asked to be in the study because they are "in the right place at the right time". It provides little opportunity to control for biases - inexpensive, accessible, and usually less time consuming to obtain than other types of samples. They are common in healthcare studies.
What is implicit order?
The objective reality beyond our own perception.