“A phenomenological study exploring UK osteopaths attitudes and beliefs in the use of Pilates for lower back pain”

“A phenomenological study exploring UK osteopaths attitudes and beliefs in the use of Pilates for lower back pain”

Question Identification & Data
Collection Methods
Learning Outcomes – Week 3
 By the end of the session and after appropriate reading and online
student-managed learning, you will be able to:
 Identify the requirements for a researchable question
 Understand the difference between aims and objectives
 Explain the concept of the Experimental and Null Hypothesis
 Construct a clear research question, and where appropriate
generate an hypothesis
 Understand the main quantitative & qualitative data collection
methods and be able to critically appraise their use.
How does this link to other lessons?
 Developing a research question requires an understanding of
the different research perspectives we looked at in Week 1, and
the different methodologies from Week 2
 To understand hypotheses an understanding of the quantitative
approach we considered in Week 1 + 2 is needed
 Data collection methods builds further on the learning from
Week 1 and 2 about research design
 You can view previous weeks’ videos at any time for a recap –
just look at the resource lists for further guidance
Developing a
Research
Question
How do we begin research?
The research question is key!
Then you need a clear and focused
research question
To begin your research project you
first need an idea
What is a researchable question?
 A question that yields facts to help:
 Solve a problem
 Produce new research
 Add to theory
 Improve practice
 A question that will provide answers that explain,
describe, identify, substantiate, predict or qualify
What is a researchable question?
 Research questions must be clear,
concise & answerable – fuzzy
questions yield fuzzy answers
 Research questions must be actionoriented
 Use a question rather than a
statement
 Go for action questions e.g.
‘what’ rather than ‘do’
 Among first year osteopathic
students at the LSO, what impact
do palpatory training exercises
have on palpation skills?
Developing your Research Question
 What do you already know?
 What do you need to find out?
 Requires broad familiarity of topic area
 Review the literature:
 Get to know the scope of the area
 Familiarise yourself with the research
 How have similar problems been studied
 Think about the approach you will take
Further resource
 There is a great summary video from Steely
Library NKU here:

(link in Resource List)
Build your own research question
 There are models that you can use to help guide you in
developing your research question
 PICO – this model defines a clinical question in terms of
the specific patient problem
 Most widely used in quantitative research
 PEO – this model defines a question in terms of the
patient and their experience of a specific problem
 Most widely used in qualitative research
Consider your perspective
 Positivist Perspective – quantitative
 Looking for facts, precision, definitions of variables and
measurement
 Interpretivist Perspective – qualitative
 Looking for meanings to different people
PICO – for quantitative research questions
The PICO Model
P Patient,
Population or
Problem
Who? How would I describe a group of patients similar to
mine?
Among…
I Intervention What? What intervention, exposure, prognostic factor am I
considering?
Does…
C Comparison to
Intervention
What is the main alternative to compare with the
intervention?
Versus…
0 Outcome Why? What can I hope to measure, improve, affect? Affect…
For more detailed information see the following YouTube video from the
University of Illinois (link in Resource List)

Example
 We want to research the impact of exercise among pre-menopausal
women with anorexia living in Southampton
 Quantitative study
 Intervention = regular physical exercise
 Comparison = no exercise
 Outcome = bone density levels
 PICO
 Among pre-menopausal women with anorexia living in
Southampton, what is the effect of regular exercise compared to
no exercise on bone density levels?
PEO – for qualitative research questions
The PEO Model
P Population
and their
Problems
Who? Who are the patients, group, community being
affected?
Among…
E Exposure Use for a specific exposure With…
0 Outcomes or
themes
What? What are you looking for? Usually there is an
element of looking at the patient’s experiences
What…
Example
 We want to research the impact of exercise among pre-menopausal
women with anorexia living in Southampton
 Qualitative study
 Phenomenological research design to understand women’s perceptions
 PEO
 What is the lived experience of anorexic women with osteoporosis
in Southampton who take regular exercise?
 A useful guide from the University Campus Suffolk is included in
Google Docs
What makes a good research question?
A checklist
 It addresses a need or a problem that you encounter as a practitioner
 You have identified a gap in the literature
 It challenges you to question your own assumptions about what you do and
what you believe
 It is researchable, meaning you are able to collect evidence that would
answer your question
 It is doable given your time and material constraints
 It inspires you and has the potential to hold your interest over a long period
 It is not too general or too narrow
 It cannot be answered ‘yes’ or ‘no’
Osteopathic Relevance
 You must be able to relate your research proposal to
osteopathic practice or theory
 This justification must be backed up with relevant literature
 Among first year osteopathic students at the LSO, what impact
do palpatory training exercises have on palpation skills?
 Read the Research Proposal Section Example on Osteopathic
Relevance
Osteopathic relevance – examples
 “Palpation is described by the General Osteopathic Council (GOsC) as ‘a
defining characteristic of osteopaths… and is considered to be one of the
primary communication channels in diagnosis, treatment and evaluation’
(GOsC, 1999 p.16). Palpation can therefore be regarded as an essential and
integral part of osteopathy, and key to clinical practice (Stone, 1999)”
 “A study by Sizer et al. (2007) concluded that there are eight motor skills
associated with manual therapy, of which two are fine motor skills, namely
fine sensorimotor and discriminate touch, which correspond to the term
‘palpation’. One of the problems of teaching these fine motor skills is that
the process is subjective and difficult to articulate (Browning, 2010). Whilst
Comeaux states that ‘there is more to the reality of contact with the patient
than a rational analysis of material tissues’ (Comeaux, 2005) p.323), a
quantifiable way of developing this faculty is nonetheless necessary for
clinical effectiveness, and a simple acceptance that experience will build the
skill is no longer appropriate within the scope of Evidence Based Medicine
(Fryer, 2008).”
Setting Aims & Objectives
 Aim
 An intention or aspiration; what you hope to achieve.
 Aims are statements of intent, written in broad terms.
 Aims set out what you hope to achieve at the end of the project.
 Objective
 A goal or a step on the way to meeting the aim; how you will achieve it.
 Objectives use specific statements which define measurable outcomes.
For example: what steps will you take to achieve the desired outcome?
Setting Aims & Objectives
 Objectives should be S.M.A.R.T.:
 Specific – be precise about what you are going to do
 Measureable – you will know when you have reached your goal
 Achievable – Don’t attempt too much
 Realistic – do you have the necessary resources e.g.: time, skills, etc.?
 Time constrained – determine when each stage needs to be completed

Aims & Objectives – Example
 Among first year osteopathic students at the LSO, what impact do
palpatory training exercises have on palpation skills?
 Aim
 The aim of this study is to assess fine discriminative and sensorimotor
palpatory accuracy after daily palpation training in a cohort of
undergraduate osteopathic students
 Objectives
 Identify a base level of palpatory accuracy
 Measure changes in palpatory accuracy subsequent to a daily, selfadministered palpatory exercise at two-week and four-week time points.
 Investigate if training compliance, age or gender affects palpatory
accuracy
 Aims
 To determine whether palpatory training exercises improve fine discriminative and sensorimotor
palpatory accuracy.
 To utilise simple training exercises to provide a cost-effective method of enhancing palpatory
sensitivity.
 Objectives
 To test the experimental hypothesis ‘palpatory training exercises improve palpatory accuracy’ against
the null hypothesis ‘palpatory training exercises do not improve palpatory accuracy’.
 To relate the outcome of the hypothesis to relevance in osteopathy.
 To utilise the student body of the LSO as the experimental group.
 To use a simple set of exercises that are accessible to all students.
 To capture palpatory accuracy in an initial exercise, teach a simple training programme to be used in
1 month timeframe and then measure palpatory accuracy again.
 To determine intra-examiner differences between pre-test, post-test and retention test to determine if
palpatory accuracy has improved.
 To determine inter-examiner differences to establish if senior years show greater initial palpatory
accuracy than junior years.
 To determine is skill transfer is achievable.
Hypothesis
Testing
The hypothesis
 An hypothesis is a specific statement of prediction
 A scientific hypothesis is an ‘educated guess’ for an observable
phenomenon
 Provides temporary explanation based on observed evidence until
it can be rejected or confirmed
 E.g. Increased chocolate intake will result in an increase in
waist circumference
 Based on previous observation and knowledge for which current
scientific theories provide no good explanation
 Must be testable
 A confirmed scientific hypothesis might become part of a theory or
it might grow to become a theory itself
The Experimental & Null Hypothesis
 To test a hypothesis we generally work with two hypotheses:
 The original hypothesis, known as the Experimental Hypothesis
(H1
)
 Increased chocolate intake will result in an increase in waist
circumference
 The Null Hypothesis (H0
), which states there will be no difference
between the variables in the Experimental Hypothesis (H1
)
 Increased chocolate intake will have no effect on waist
circumference
The Experimental Hypothesis
 States that a population parameter …
 …is not equal to
 OR … is greater than
 OR … is less than a claimed value
 E.g. LSO students will have more grey hairs at the end of Year 4
compared with the beginning of Year 1
 Experimental Hypothesis = ‘true difference’ or ‘true effect’
Null Hypothesis
 NOT the opposite of the Experimental Hypothesis
 It is an hypothesis that says there is no statistical significance
between the two variables in the hypothesis
 The Null Hypothesis is considered to be true until evidence indicates
otherwise
 It is the Null Hypothesis that the researcher is trying to disprove
 We test the Null Hypothesis rather than the Experimental Hypothesis
as we can prove a negative, but we can’t prove a positive
 If the Null Hypothesis is false we can ‘reject’ it
Confidence in Hypothesis Acceptance
 How confident are we in rejecting the Null Hypothesis?
 We can never be completely sure that either hypothesis is correct –
because we are not analysing the entire population but only a sample
of it
 So we work with probabilities to show that the results we obtained did
NOT occur by chance!
Probability
 If the probability that our result occurs by chance decreases, the
confidence grows that the Null Hypothesis can be rejected more
safely
 To decrease the probability that our results occur by chance, we make
sure we analyse a large enough sample of a population
Fisher’s Tea Tasting Lady
 A lady claims she can taste whether tea is prepared by adding milk
before or after pouring tea into a cup
Fisher’s Tasting Tea Lady
 Experimental Hypothesis: Lady can determine tea preparation by tea
tasting
 Null Hypothesis: Lady cannot determine tea preparation by tea tasting
 Experiment: the lady tastes a number of cups of tea with or without
milk added first and observe if she identifies the right order of milk
first / milk after
Probability of tasting tea correctly
 If we put two cups of tea in front of the lady…
 She has a 50% chance of guessing correctly
 If she identified the tea preparation correctly, we would not be very
confident that she can really tell the difference in tea preparation –
most of us could have guessed it by chance (50/50)
Probability of tasting tea correctly
 If we put 6 cups of tea in front of the lady, she has a smaller chance
of guessing the order of tea preparation correctly
 The probability that she picked the right order of cups by chance is 1
in 20 – as there are 20 different ways the cups could be ordered
 If she identified the tea preparation correctly, we would be very
confident that she really could tell the difference as the chance that
she guessed correctly is only 5%
The p-value
 In biological sciences, we usually reject a Null Hypothesis if there is
only a 5% or 1% chance that we wrongfully reject it and instead accept
the Experimental Hypothesis
 The 5% and 1% thresholds are expressed as:
 a=0.01 for 1%
 a=0.05 for 5%
 Statistical tests can now calculate the smallest value, called the pvalue (probability value), at which the Null Hypothesis can be safely
rejected
 Reject the Null Hypothesis if the p-value < a
 Fail to reject the Null Hypothesis if the p-value > or = a
Tea tasting and p-value
 Experimental Hypothesis: Lady can determine tea preparation by tea
tasting
 Null Hypothesis: Lady cannot determine tea preparation by tea tasting
 Result: lady has correctly identified all 6 cups
 Probability of guessing correctly: 5%
 We reject the Null Hypothesis and accept the Experimental Hypothesis
 The probability that we falsely rejected the Null Hypothesis and
accepted the Experimental Hypothesis is only 5%
 p=<0.05

Quantitative
Data Collection
Research Perspectives or Paradigms
 Positivism
 Reality exists independently ‘out
there’
 Objectivity & absolute truth
 Studies focus on measurable
attributes – facts and figures
 Deductive
 Findings generalisable
 Quantitative approach
 Interpretivism
 Reality is constructed through the
meanings of individuals
 Subjectivity & relative truth
 Studies focus on people’s beliefs
and lived experiences
 Inductive
 Findings not generalisable
 Qualitative approach
Quantitative
Research
Uses numbers as the unit of analysis
Associated with large-scale studies
Associated with analysing specific variables
Associated with data analysis after data
collection
Associated with researcher detachment
Data collection methods
 Data collection is the gathering of information
 A data collection method is the instrument or tool used to gather the
information
Quantitative data collection methods
 Experiments
 Questionnaire surveys
 Structured observation
Experiment
 An experiment is an investigation carried out under controlled
conditions to examine the relationship between specific factors
 Isolates individual variables and measures their effect
 Requires:
 Controls: the manipulation of variables by the researcher
 Empirical observation and measurement: detailed and precise
 Identification of causal factors: measurement of the factor that
causes an observed outcome
Types of experiment
 Laboratory experiments
 Experiments carried out at a location that is set out purposefully
to aid the collection of data
 Field experiments:
 Carried out outside the laboratory in places such as factories,
hospitals, schools
 Randomized Controlled Trials (RCTs)
 A type of field experiment in which the effect of a specific
‘treatment’ is compared between experimental and control groups
 Random allocation of participants to each group
Experiment
Advantages
 Credibility
 Repeatable
 Precision
 Cause & explanations
Disadvantages
 Artificial settings
 Challenge of controlling
variables
 May miss indirect causes
 Issues with random sampling
Questionnaire Survey
 A survey asks participants questions to provide a comprehensive view
of something at a specific point in time
 Surveys aim for wide and inclusive coverage
 Survey may be:
 In-person, internet, telephone or face-to-face interview, postal
 A questionnaire is a written list of questions designed to collect
information for subsequent data analysis
Questionnaires
Advantages
 Economical
 Relatively easy to organise
 Standardised delivery of
questions
 Data processing ease
 May facilitate accessibility
Disadvantages
 Pre-coded questions can be
frustrating
 Impose a structure on answers
 Challenge of truthfulness
Qualitative
Data Collection
Research Perspectives or Paradigms
Positivism
 Reality exists independently ‘out
there’
 Objectivity & absolute truth
 Studies focus on measurable
attributes – facts and figures
 Deductive
 Findings generalisable
 Quantitative approach
Interpretivism
 Reality is constructed through the
meanings of individuals
 Subjectivity & relative truth
 Studies focus on people’s beliefs
and lived experiences
 Inductive
 Findings not generalisable
 Qualitative approach
Qualitative
Research
Uses words or visual images as the unit of analysis
Associated with small-scale studies
Associated with a broader perspective
Associated with data analysis during data collection
Associated with researcher involvement
Uses inductive reasoning
Data collection methods
 Data collection is the gathering of information
 A data collection method is the instrument or tool used to gather the
information
Qualitative data collection methods
 Interviews
 Focus Groups
 Observation
 Naturally occurring data
Interviews
 Interviews collect data by the researcher asking the participant
questions
 Different from conversations
 Agenda for discussion set by researcher
 Interviewee’s words can be used as research data
 Can be used to explore thoughts, feelings, experiences
 In-depth discussion, usually around 1 hour
 Audio-recorded and transcribed
 Requires researcher reflexivity
Types of interviews
 Structured quantitative interviews
 Format of questions tightly controlled – similar to a questionnaire
 Pre-determined list of questions
 Useful for large-scale projects
 Semi-structured interviews
 List of issues to be addressed and questions to be answered
 Researcher flexible in ordering of questions and allows interviewee to
develop a topic
 Open-ended questions
 Unstructured interviews
 Non-directive
 Researcher introduces a topic and interviewee allowed to develop the
discussion
Interviews
Advantages
 Depth (richness) of information
 Insights from ‘key’ informants
 Persuasive evidence, capturing
own words
 Flexibility in design
 Increased dependability of data
 Higher response rate
 Participant enjoyment
Disadvantages
 Large amounts of data
 Complex to analyse
 Possibility of ‘interviewer
effect’
 Can be time-consuming
 Data protection issues
 Requires skill of interviewer
 Reflexivity requires practice
Types of interviews
 1:1 interviews
 Focus groups
Focus Groups
 Small groups of people (6-10) brought together by a ‘facilitator’ (the
researcher) to explore attitudes, beliefs, perceptions of a specific
topic
 Uses group interaction to generate data
 Useful for identifying and clarifying a range of views
 Usually audio-recorded and transcribed
Focus group – disadvantages
 Challenge of group dynamics
 Ensuring everyone has a voice
 Issues in capturing action
 Difficulties with transcribing
 Provide collective rather than individual phenomena
 Do not reveal consensus views, rather divergence of opinion and
recurrence of themes across groups
Observation
 Observation is carried out by a researcher watching what a person
does
 Carried out ‘in the field’ – in real life situations
 Researcher involvement ranges from ‘participant observer’ to
‘complete observer’
 Fieldnotes are recorded to describe the observation, and subsequently
analysed and interpreted
 Requires researcher reflexivity to be aware of own role in the
research
Observation
Advantages
 First-hand data
 Little equipment required
 Deep insights
 Context rich
Disadvantages
 Researcher perception – ‘going
native’
 Reflexivity requires practice
 Challenge of access
 Ethical issues if covert research
Naturally Occurring Data
 Data, often in textual, documentary or archival form, produced
without intervention by the researcher
 E.g. diaries, social network group posts etc
Designing
questionnaires
and interview
questions
Common research methods
 Questionnaire
 Tend to be used for quantitative surveys
 Semi-structured interview
 Tend to be used for qualitative research
 Both require the researcher to develop questions … but in different
ways
Designing a questionnaire
1. What do you want to find out?
 Look at your research question and write a list of the aspects you want to
find out
 Consider who your target population is
2. Write questions
 A questionnaire question is known as an item
 Each item has various response options
 For each aspect you want to find out about there may be several items
3. Desk-check
 Proof read for errors
4. Pilot survey
 Test your survey for problems!
Questionnaire items – checklist
Items should be:
 Short & simple
 Avoid unfamiliar words
 Neutral
 Avoid extreme wording (never, always)
 Avoid biased terms
 Answerable
 Can everyone answer?
 Unambiguous
 Avoid double-barrelled questions
 Avoid vague questions
 Consistent
 Should be an indicator for the aspect you are measuring
Designing a semi-structured interview
 Differ from quantitative, survey-based questions:
 Seeks a more in-depth, nuanced response
 Acknowledges that there is no ‘one-size fits all’ for people’s
experiences
 Questions are not fixed items but provide a guide – extra
questions can be added about an unexpected but relevant area if
it emerges
Designing the interview guide
1. Start by developing specific questions that will help you collect data
to answer your research question
 Use open-ended questions (that cannot be answered by a yes or
no)
 Use words like: why, how, what, describe, explain
 Avoid leading questions
2. For each specific question, add a list of prompt questions to help you
elicit deeper information
 These questions remind you of more areas of detail that might be
covered
Good Practice – Interviews
 Prompts: Use if interviewee is struggling to answer
 Offer some examples, repeat the last few words spoken by the interviewee
 Probes: Use to find out more in-depth data
 Ask for an example
 Ask for clarification
 Ask for more details
 Check: summarise main thoughts e.g. So if I understand you correctly…
Tips for carrying out an interview
 Introduce yourself, explain the research and get the participant’s consent. If any
recording devices are being used, point them out to the participant and make sure
they’re working.
 Ask warm-up or demographic questions first; then, using the interview guide,
move on to more focused questions
 Be flexible with the order
 Know your guide backwards – this will help conversational flow
 Try not to interrupt participants; make a note and come back to the idea later
 Use non-verbal encouragement e.g. gentle head nods or “um-hm”
 If a participant gives an answer relating to a question you have not yet asked,
record this in your notes and avoid repeating the question later
 Try to avoid tangents. Spend time on key factors, including what the participant is
interested in speaking about
 Make sure interviewee is doing most of the talking
 If time permits, ask the participant if there is anything else they’d like to share
Student-managed learning
1) Research:
 Read Osteopathic relevance section Research Proposal Guide
 Research and read literature pertaining to your research idea
 Make notes and save articles
2) Review understanding:
 Read ‘HYPOTHESIS / P VALUES / PROBABILITY ETC
 Makes notes as appropriate
3) Recall:
 Try explaining probability and p-value
4) Reflect:
 Consider the research you have read about the topic you are interested in, and if
your research idea is still appropriate or whether it needs adapting / changing
References
 Bruce, N., Pope, D., & Stanistreet, D., 2018. Quantitative methods for health research: a
practical interactive guide to epidemiology and statistics, 2nd Edn. Chichester: John Wiley &
Sons
 Creswell, J. W. & Poth, C.N., 2017 Qualitative inquiry and research design: choosing among
five approaches, 4th ed, London: Sage
 Green, J., & Thorogood, N., 2018. Qualitative methods for health research. 4th Edn. London:
Sage
 Lewith, G. T., Jonas, W.B. and Walach, H. 2010. Clinical research in complementary
therapies: principles, problems and solutions. 2nd Edn. Edinburgh: Churchill Livingstone.
 Moule, P., Aveyard, H., Godman, M., 2017. Nursing research: An introduction. London: Sage.
 Passer, M.W., 2017. Research methods: Concepts and connections. 2nd Edn. New York:
Macmillan Education.
 Robson, C. and McCarten, K., 2016. Real world research. 4th Edn. Chichester: John Wiley &
Sons.
Any questions?