CH5 Qualitative and mixed methods approach

Step 1 role qualitative methods

In a systems approach qualitative methods can play a role in step (1) delineating the system \n by identifying the system’s components, relations, in- and outputs and the larger context. \n Researchers can do this by themselves based on the literature, but qualitative methods can \n help to identify the problem together with stakeholders, those people who play a role in the \n system. These people can help deepen our understanding of what is happening and why it is \n happening. As a result more specific (partial) research questions can be developed.

Step 2 role qualitative methods

In step (2) of the systems approach qualitative methods can be used to collect data to \n answer the (partial) research questions. They can also be used to collect data together with \n stakeholders and/or to get feedback on results to help interpret the findings.

Step 3 role qualitative methods

\n In step (3) \n qualitative methods can help link and integrate different knowledges (scientific knowledge \n integrated with for example knowledge based on experience and knowledge based on \n tradition). Here you can also use qualitative methods to communicate your findings. Overall, \n qualitative methods can play an important role in the study of animal production systems \n

3 parts of study desing

1. How often do you need to meet your participants to answer your research question?
If you interact with participants only once we call this a cross-sectional study design. If
you meet them two times, it is referred to a before and after study design and if you
meet them three or more times it is considered a longitudinal study design.
2. What is the reference period?
The reference period relates to whether you want to know about the situation
retrospectively (looking back to the past or what has happened so far) or
prospectively (looking forwards to the future). In some research it is relevant to
include both reference periods.
3. And what is the nature of the investigation? This relates to whether the research involves an intervention from the researcher to test for causality. In experimental research they set up an experiment, an intervention, to see if a change or difference occurs. In non-experimental research the research starts from the situation with or without changes or effects to then try an understand what the causes might be

Interact with participants only once

cross-sectional study

meet with participants twice

before and after study design

meet with participants three or more times

longitudinal study design

retropescrively

looking back to the past or what has happened os far

prospectively

looking forwards to the future

Experimental research

Experimetn set up

Non-experimental research

research starts from the situation with or without changes or effect to then try and understand what the causes might be

Sampling is made up of

Sample criteria

Sample size

Sample approach

Sample criteria

On what do you select them

Sample size

How many particiants of the population is neded to engage with to be able to answer your research question

Sample aproach

Probability

Non-probability

probability sampling or random samping

Everyone has an equal chance to be invited to participate in the research

random sampling

Take the full list of the population and do a simple random selection by allocating random numbers

Systematic random sampling

you consciously choose a systematic approach

stratified random sampling

A stratified random sample is a sample in which the full list \n of the population is subdivided into smaller lists, based on certain criteria. The relative \n size of each of these subgroups in comparison to the full list is also reflected in the \n number of people that are subsequently randomly sampled from within these \n subgroups. \n

Cluster sampling

in this approach the population is subdivided into clusters, for \n example in administrative regions, such as municipalities.

Types of random sampling

Random sampl

Systematic random sample

Stratified random sample

Cluster sampling

Non-probability sampling

Not everyone has an equal chance to participate

Types of non-probability sampling

Purposeive sampling

Quota sampling

Snowball sampling

Convenience samplingP

Purposive sampling

Purposive sampling is one approach in which the sample is based on a specific \n purpose: particular criteria that are relevant to the research question. Criteria can be \n based on wanting to collect data from different groups, or from the same type of \n people.

quota sampling

it is based on subdividing a larger \n population into subgroups and using the relative size as guidance for sample size. \n However, in quota sampling the engagement stops once the quota is fulfilled \n

volunteer sampling, includes snowball sampling

approach depends on people you know and other people they know. the sample is then taken from a personal network of people who know eachoterh

convenience sampling

Convenience: recruiting people you know or who you know are willing to participate

Types of questions

Ranking questions (e.g. rank these items in order of....(e.g. importance)) \n • Likert scale questions (e.g. How much do you agree or disagree with the following \n statement(s)? strongly disagree < -- > strongly agree) \n • Structured yes/no questions (e.g. Are you...; Are you in favour of...; ) \n • Multiple choice questions \n • Open questions (e.g. Describe...; name three things...; what is needed to ...

Challenges related to questions

Length of the survey or interview

Ambiguous phrasing → When term is not explained

Cocktail questions → Multiple questions are asked at the same time but only one answer can be given

Leading questions → lead people to answer in a certain way

Design of questions in term of the answer options

response bias

an answer does not reflect patticipants attitued or perspectives

halo or horns effect

when an intitial positive or negative impression affects the response to antoher question

One way to counteract this effect is by using software that can suffle the order of questions between participants, but mostly important is recognising wheter question smight result in bias

bandwagon bias

show the responses of other participants as part of the interview or survey. participants might follow the answers of other participants

Analytical objectives for both qualitative and quantitative data can include

characterizing the variables

analysing the relationships between variables

explaining and predicting variables

making comparisons

steps involved in analyses with different types of variables

recognize data types (nominal, ordinal, interval, ratio)

identify correct statistical test

test wheter assumptions associated with the test are mest

run the tests and interpret the results

aim of thematic analysis

identify patterns within and across data in relation to participiants lived experience, views and perspectives, and behaviour and practices

Phases of thematic analysis

Familiarizing yourself with your data

Generating initial codes

Searching for themes

Reviewing themes

Defining and naming themes

Producing the report

Summary

This chapter has introduced the steps involved when qualitative methods can play a role \n across the systems approach to answer a research question. In a systems approach \n qualitative methods can play an important role in helping to define the system, context and \n problem; to collect data to answer research questions, and to interpret results. A mixed \n methods approach involves both qualitative and quantitative methods. It offered a more \n detailed description of surveys and interviews as research methods and of thematic analysis \n to analyse qualitative data