Research Methodologies Flashcards
What is qualitative research?
Qualitative research explores information and non-numerical data. Analysis of qualitative research can examine how people might connect meaning to their experiences or emotions.
Pros and cons of qualitative research
Pros: Explores attitudes and behaviours, encourages discussion, flexible, can explain something which numbers alone may not be able to
Cons: Sample sizes can be small (may not be a true reflection) and may have bias, lack of privacy, skilled moderator or not
What is quantitative research?
Quantitative research uses numbers to measure data. Analysis of statistics can help find connections and meaning in the data.
Pros and cons of quantitative research
Pros: Can use larger sample sizes, accurate and impartial, fast, based on scientific data in my instance which is verified
Cons: Research typically not in a participants normal environment, un-able to follow up or expand on points
What is a mixed method?
Mixed methods research integrates both qualitative and quantitative measures.
It is a holistic approach to combining statistical data with contextualised insights
Pros and cons of mixed method approach
Pros: Mixing data sets can give you a better view of the overall picture, it can strengthen findings (triangulation), helps researchers develop their skills
Triangulation is the confirmation of results by different methods
Cons: A lot of resource and skills required, not suitable for all areas
What is research design?
A research design is a high-level strategic plan which defines how you plan to answer your primary research question. It defines tools, techniques and methods to implement the plan.
What are research methods?
Research methods are processes used to collect data. Research method types include interviews, focus groups, surveys, observation
What are qualitative research methods?
Interviews, focus groups, observations, document analysis, oral history
What are quantitative research methods?
Surveys, questionnaires, observation, experiments, numerical screening
What is the Mental Capacity Act 2005?
Participants need a presumption of clarity to participate
What’s the ICO?
Independent Commissioner’s Office (ICO) exists to provide individuals and organisations with support and empowerment through information. They police the General Data Act and GDPR!
What is the Terrorism Act 2006?
I am aware that security-sensitive should be kept off personal laptops and on specifically designed servers supervised by ethics officers
Examples include research commissioned by the military, EU security call, concerns terrorist groups or involves the acquisition of security clearances
Who creates and monitors research ethics?
Faculty Ethics Committee (FEC) or the Subject Ethics Panel (SEP)
What is Innovate UK’s view on ethics?
Supports the UKRI ethics which include upholding the rights and dignity of participants, base partnerships on mutual respect, research conducted with integrity and transparency, participants should be informed
Who is responsible for ethics?
Ethics is a collective responsibility of the researchers. Researchers are encouraged to undertake self critical ethical reflection throughout duration of a project.
Tell us about a time when you made a methodology
Scope 3 example
Tell us how you choose which research method to use:
It depends on the result you are pursuing. If I wish to know gain statistical clarity or a quantified answer, I would choose a quantitative approach with metrics I can measure.
If I wished to know feelings or if the topic is intangible or something which could explain what numbers cannot, then I would use a qualitative approach.
For the carbon metrics for the Regenerative Twin, we wished to measure this against 4 ‘climate change mitigation’ numerical metrics. This meant the research type was quantitative as we wished to see if it were achieved.
Example of a sense check of data I’ve collected
Charred timber cladding for the regenerative twin – I might sense check this against a different EPD for charred timber cladding to triangulate the figures
What is the difference between qualitative and quantitative research methods? When is it best to use one?
Quantitative research involves numerical metrics to measure data whereas qualitative research methods involve gathering behavioural data and reasoning.
The use of each depends on the project type. If measuring or analysing numerical data to achieve something, then I’d use quantitative but if wishing to understand the meaning, then I’d use qualitative.
What is the difference between primary and secondary data? Which is more reliable?
Primary data is first-hand, raw collected data whereas secondary data has come from another source.
Primary data is more credible. The reliability of secondary data depends on the source. However, a good piece of research includes both primary and secondary.
Example of a time using qualitative methods
Gathering feedback from the champions in London (Forms)
Strengths of this approach: Time (and cost) efficient, quantitative and qualitative as it had room for text, can be translated into graphics to digest, provide actions, good for 45+ employees, uniformity with standardised responses, can be anonymous, can be replicated with ease
Limitations of this approach: Sometimes lost in translation/opinions needs explaining, susceptible to ‘response bias’ if not anonymous, people can ignore them, questions could be misinterpreted, cannot capture complex information
Example of a time using quantitative methods
Using CarboniCa on the Regenerative Twin project
Pros: Accurate data, presents data graphically for easy interpretation, does not need to be a consultant filling it out, easy optioneering
Cons: Sometimes the incorrect unit was provided and a calculation would be needed for m3, only 1 person filling it out
Example of a time where you have recommended improvements for a methodology/data collection:
EPD gate system OR champions survey
How do you project manage R&D for CarboniCa?
I project managed CarboniCa R&D by having a tracker, aligning objectives to the client’s strategy, and having weekly calls
* I had a CarboniCa development meeting on Thursday afternoons with MSES and ConstructSys and then a CarboniCa weekly catch up on Friday afternoon for queries
* I had a Teams channel for all stakeholders called CarboniCa R&D
* Have a group titled ‘Refining CarboniCa’ who are advanced users and provide feedback on upcoming new updates. This group meets quarterly
* Present at the carbon champions forum x3 a year on upcoming changes
Tell us about the EPD gate system and how data is transferred?
Users were adding EPDs to their projects and they were going straight into the central EPD library within the carbo factor database. However, sometimes these contained inaccuracies.
I discussed and advised MSES and ConstructSys to create a mechanism whereby user added EPDs were sent to a gate, rather than into the central database.
At this ‘gate’, they were sense checked by MSC then the system would alert MSES who would do a secondary review and release the data into the EPD library.
What data sources did you use to calculate the Recommendations section within CarboniCa?
CarboniCa’s carbon factor database which comprises of Bath ICE data, government convergence factors and an EPD library
What data sources did you use to calculate the Low Carbon Flooring guide figures per m2?
CarboniCa’s carbon factor database which comprises of Bath ICE data, government convergence factors and an EPD library
How did you establish the client requirements for the ICCE project?
These the requirements of the Innovate UK funding which we evolved into the client brief
Client wanted to develop CarboniCa into a tool for Education so I set out to find with the MoSCoW method what that would need to look like
What data did you provide the cohort with in the ICCE project?
Workshop 1 – Overview on CarboniCa
Workshop 2 – Data for a typical steel frame secondary school (MDS)
Workshop 4 – App development & tutorial data
Strengths and limitations of the MoSCoW method. Why was this selected?
Strengths: Facilitates discussion, good for sample size we had, interactive (post it notes on wall), simple approach, easy to use, selected as it is an efficient prioritisation method.
Limitations: Does not prioritise exactly what should come first if there are a lot of ‘Must haves’, sometimes the same item is in different categories (no qualitative understanding of why), can be team bias but we used it individually
How has the workshop informed the proof of concept?
The 6 workshops hosted at NTU have informed the proof of concept by prioritising what students and academics would need in a carbon tool. This includes:
* Easy navigation and clear graphics
* Tutorials to explain data entry, and how to interpret results
* Pre-populated examples
* A vast carbon factor database
What industry data have you collected?
EPDs, LCAs, reports e.g. IPCC, TCFD
Talk me through research methodologies you have used
- MoSCoW example - NTU
- Delphi method – NTU
- Questionnaires – Carbon champions forum
What is the Delphi method?
The Delphi method is a structured communication technique whereby forecasting is done with a panel of experts.
Example – sustainability module at NTU. The proposed module was discussed by the viewpoint of 5 departments. The process will review all current resources, agree on shared content, evaluate it. The needs of the module in the curriculum were discussed e.g. themes would be awareness, case study examples, theories and concepts, etc
How have you provided advice based on research?
- Creation of the client aspirations part after questionnaire
- Advice to tailor tutorials a certain way after MoSCow feedback e.g background, data interpretation and customisation
Tell me about CarboniCa and how new data is added?
Two fold approach to this:
1. User can add EPDs to their own project
2. Batch upload of new carbon factors in every release update – do 3x per year
What problems have you encountered with CarboniCa?
- Being slow
- Incorrect calculations
- Perception as being difficult and an add on
What is a sensitivity analysis?
A sensitivity analysis determines how different values of an independent variable can affect a particular dependent variable under a set of given assumptions. It can help you make informed choices
What is scenario analysis?
A scenario analysis looks at the potential future business impacts and events in relation to one scenario to see how this is impacted and how alternatives outcomes could happen. Example – TCFD scenario analysis
Increasing up front practical carbon to reduce whole life – explain this please
You often need to invest in up front embodied carbon to reduce whole life. For example, building services might be higher in up front embodied carbon but more efficient so they reduce whole life carbon
What is data handling?
Data handling is the process of ensuring that research data is stored, archived or disposed of in a safe and secure way
What are data handling techniques?
Data handling refers to using tables to organise information in rows and columns, making structured data easier to understand.
What are data manipulation techniques?
- Gather data from several sources
- Organise and purify the data
- Combine datasets and eliminate redundancies
- Utilise data analysis to discover information
What is data manipulation?
Data manipulation is the process of organising data to make it more understandable. For example, data might be alphabetically for easy comprehension.
What are the steps of data manipulation?
- Understand the data
- Data cleaning – pre-process to remove inconsistences, missing values, etc
- Data transformation
- Filtering and selection
- Aggregation and summarisation – like creating a summary table
- Data visualisation – visualise data to identify trends, patterns, outliers, etc
- Modelling – select appropriate algorithms to train the model to manipulate data e.g NTU MATLAB AI project
- Iteration – repeat the process to achieve the goals
- Testing – ensures reliable results
- Consider ethical considerations throughout – privacy, biases, etc
What are data manipulation tools/IT support?
Excel (VLOOKUP and pivot tables), the Pandas library in Python to analyse and manipulate data, R which is a statistical programming language that can manipulate data
Example – in Excel, I can filter rows and columns to see different trends e.g. who is top for 10tC total, or 10tC per £100k project value or I can create a pivot table
How do you deal with sensitive information?
- Sign relevant documentation e.g NDA if external
- Keep in secure location on professional laptop only
- Examples - password protected EPD spreadsheet to avoid human error, controlling access
Relevant legislation regarding handling and using data:
- Data Protection Act 2018, GDPR for personal data
- The Mental Capacity Act 2005 – participants need a presumption of clarity to participant
- The Terrorism Act (2006) & Security Sensitive Research – kept off personal laptops, ensure safe servers, etc
