Introducing 

Prezi AI.

Your new presentation assistant.

Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.

Loading…
Transcript

MPA620

Date: February 24th 2021:

Master of Public Administration Program

AGENDA

AGENDA

Selecting your research methods

Interviews and focus groups

Designs and approaches to data analysis

Group discussions (team work)

Qualitative methods

Main qualitative methods

Selecting your research methods

Requires a numerical description

Quantitative methods

There is no necessity to report results and tie them to specific socio-cultural context

When the replicability is important

Are appropriate when...

When researcher wants to expand the results beyond contexts and compare populations

When indicators have already been developed to report given realities

In other words, when the goal is EXPLANATORY

The theme is new, not familiar

Qualitative methods

Exploratory research: important concepts and variables are unknown

Need for a thick description: the goal is not to quantify but interpret behavior or phenomena (depth, meaning)

Are appropriate when...

Looking for sense (not frequency)

More flexibility in the approach to make unexpected findings or explore in depth particular themes

Research aims at a detailed understanding or particular dimensions, situations or events

Qualitative research is... more inductive

Which means... ?

Making sense from a specific case or cases and outline a process, a pattern that would apply more generally to other cases (the opposite of deductive)

This research methodology can help encompass the needs of a population, can explain and interpret the results of a quantitative inquiry

The object of inquiry can be individuals, specific populations, or the general population.

Best appropriate to develop hypothesis to test in subsequent inquiries.

Several techniques can be used when using a qualitative approach

Strategies associated with the Qualitative Approach

Ethnographic research: to study an intact cultural group in a natural setting over prolonged periods

Grounded theory:deriving a general theory from the grounded views of participants

Case studies: detailed exploration of a case bounded in time through a variety of method. (Cases can be: program, event or activity, an individual or a process).

Phenomenological research: identification of "the essence" of experiences concerning a phenomenon as described by the study's participants.

Narratives: to study the lives of individuals to provide stories about them. Combination of experiences

Knowledge claims, strategies of inquiry and Methods

Methods

Semi-structured

In depth

Overt

Undercover

Individual

Group (semi-structured)

Direct<

Indirect

Interviews

Observations

Participating observer

Non participating observer

Conversationalanalysis

Documentary analysis

COMPARING QUALI, QUANTI & MIXED APPROACHES

the weak

Criteria for selecting methods

Qualitative is constructivist, interpretative.

  • The main goal is to develop themes from the data.
  • The researcher collects open-ended emerging data, multiple meanings of individual perspectives.
  • She or he collaborates with the participant and seeks to understand (interpret) social and historically constituted constructs
  • Observation of behavior, ethnographic designs and emancipatory research perspectives are qualitative in approach.
  • The researcher seeks to establish the meaning from the view of the participant.
  • This means identifying a culture-sharing group and seeking knowledge about shared patterns of attitudes or behaviors.

Criteria for selecting methods

Quantitative is experimental, quasi-experimental:

  • Explanatory - Inferential (effect) are quantitative in approach.
  • Sets out to test an existing theory.
  • It involves hypothesis, and rates attitudes and behaviors.
  • The researcher uses a "cause and effect" thinking; reduces the inquiry to specific variables and hypothesis.
  • Experiments and surveys are the primary methods for collecting data
  • In such a design, there is no procedural flexibility; only predetermined instruments.

Criteria for selecting methods

Mixed methods are pragmatic by nature.

  • This approach triangulates: looks at the same phenomena from multiple angles, is problem-centered and pluralistic:
  • Sometimes, the two approaches (quanti and quali) are used simultaneously, but usually, both approaches are used in sequence: the study begins with a broad survey in order to generalize results to a population and then focuses on detailed, qualitative open-ended interviews to collect detailed views from participants.

Criteria for selecting a design - remember...

If it is measurement oriented, it is quanti

If it is meaning oriented, it is quali

If it is problem-oriented (pragmatic) the most appropriate design is probably mixed method (sequential or simultaneous)

Gathering qualitative data

What is the purpose of an interview?

And of a focus group?

Interview and focus group

Interviews

Interviews

Standardized procedure aimed at accessing the conceptual framework of the interviewee.

- It has varying degrees of structuredness

When are in-depth interviews appropriate?

When are in-depth interviews appropriate?

When exploring a complex topic, with a well informed population

When the topic is sensitive (intimately related to someone's path, for example - women in prison and factors affecting their incarceration)

When it is hard to gather data from the geographically dispersed participants <

When there is a risk of peer's contamination (in other words, social dynamics may influence measurement)

Content analysis

Deductive content analysis

The principles:

1- Initial reading matrix

(selection of thematic to lead the analysis of narratives)

2- Translation of the reading matrix in a series of codes

3- Structuring of the qualitative data according to a series of variables (codes)

4- Systematic coding and analysis of coded materials

5- Statistical analysis (distribution and co-occurrence of the codes in particular)

(Bardin, 2006)

My interviews are done! (Now what...?)

Focus groups

Focus Group

  • Usually: : 4 to 8 participants
  • Predetermined selection criteria
  • Homogeneity within the group
  • Subgroups can be constituted (by gender, age, etc.)

The discussions usually last until no new information comes up.

Average 1 hour, 1 1/2 hour.

Focus groups are appropriate when...

Focus groups are appropriate when...

Unit of analysis is the group.

To explore attitudes, perception, impressions (intra-group dynamics)

Assess concepts, define an approach (tailor an intervention, a message)

It also helps the process of identifying and defining problems.

How to lead a focus group?

Questions and thematic are pre-determined

The interview guide comprises a series of open questions

The participants discuss among themselves

Possibly, record the discussion

Paper or computer transcripts

Coding...

Qualitative data must be analyzed systematically.

You can code texts using color pencil, highlighter or word (computer)

Codes essentially help you structure for different emerging ideas.

Coding...

For focus groups, you must analyze each group's discussions (not by individual) and use multiple sources to make connections (triangulate)

Look for intra-case and cross case patterns.

Designs and approaches to data analysis

Analysing qualitative data

Atlas.ti

Anselm Strauss &Juliet Corbin helped develop it.

Strengths:

  • Simple, intuitive functioning, and easy to access.
  • Coding is clearly presented, exhaustive, and adapted to inductive reasoning

Weaknesses:

  • There are not a lot of possibility for structuring information
  • The integrated quantitative tool is weak.
  • The interface in unappealing

Two software

NVivo

Strengths:

  • Appealing interface.
  • Perfect for content analysis
  • Vast possibility for informational structuring
  • Integrated quantitative tools

Weaknesses:

  • The codes you use are somewhat "hidden" which creates
  • limitations
  • Not as easy as NVivo to master
  • Costly

Principles for analyzing qualitative data

Quantitative research uses and reports predetermined categories : descriptive statistics, charts, correlation coefficients, etc.

In qualitative analysis, the process is quite different.

In a qualitative approach, the instrument and classification are typically more flexible.

There are 3 types of coding strategies.

- Purely inductive (emerging patterns only)

- Semi structured : You have predefined categories, but add to them

- Structured: deductive

Some difficulties to expect

When choosing an unstructured coding strategy (inductive), the establishment of the codes happens during the analysis. Each time a category emerges, you must go back to your previously analyzed transcripts and re-code your material.

When deciding what the categories will be before starting to analyze (deductive approach) you will be tempted to add to your initial grid whose coherence may be challenged by the data.

Most researchers use a semi-structured coding strategy because of time constraints. This way, your analytical matrix remains flexible without having to start categorization from scratch .

An example

We did’nt you know, to begin the operations, you know, because we had the regulations on our side, you know. We wanted the community to support us, you know… as much as they could and so there were things that that the town asked for that, in a typical railroad operations, you know, might not be something that a railroad would agree to. So those items were things like: ensuring that you have you know, two people, on the trains that operate through the town of Lac-Mégantic. Ok. No problem. Am.. they wanted us to operate through the town at 10 miles an hour. Ok, that’s fine, and that’s important to the community, that’s what we’ll do.

A: hum hum.

RR: Am.. They.. It was important for them, to not have our freight trains stopping in town and doing, you know, a lot of train operations, moving cars back and forth from the other. Am.. So we’ve you know, we moved some tracks and we moved, you know, some of the operations.. We’ve found a different way to do it, you know,

A: hum.

RR: Just tried to be a good new neighbour. But back to the, back to your original question, you know, a lot, most of the regulatory changes that have been put in place am, you know, and they were put in place almost a year and a half after the accident occurred. Am. The most significant, is the train securement rules,

A: hum hum.

RR: And, you know, the number of hand brakes that you have to tie, and the testing and the things that go into that. Well, Jeremy mentioned that we have made some operating rule changes on day one. And those were changes that we made before the regulators told us..,

(private sector respondent - executive level)

Another example

The SMS regulation was too vague. (...) Simple example, the regulation said: "You had to have a method to analyze risks". When you do your audit, you ask: "Do you have a method to analyze risk". You show it to me. You meet regulation... You meet it! Even of it is crumbling, [the regulator] can't do anything. (...) An 18 pages legal advice defining implementation was telling [those in charge of the regulation enforcement they ]did not have the right to go into the implementation level, that [TC] did not have jurisdiction over this. Either you do or you don't! (Interview #27, senior government manager with prior private sector senior management experience)

And another...

"With the absence of statistics demonstrating [to the Department of Transport Canada] that there was a problem, in the absence of a database allowing [TC] to look at the compliance rate (...) it has been difficult for [TC] to see there was any problem with MMA or another railroader. You rely on the intelligence you have on the field. (...) Everyone who has been working on computer applications know how it is always a little and that it is always cost a little more...". (Interview #27)

WHAT'S NEXT?

Deadline for the final paper ("project proposal"

project) is 10/14/2020

General Reminders

Project proposal’s components:

• A “problem statement” description of the chosen topic and its justification (pertinence).

• A short literature review explaining the core component of the research problem and the limitations in existing research (that is a section with your critique and analysis of the literature’s shortcomings).

• A specific problem (gap).

• A statement regarding what is the research question (specific question) and research aim.

• A description of the research methodology: research’s approach, design, dimensions and strategies.

• Specification of dimensions, variables, and hypothesis (if applicable).

• Description of data collection methods and analytical methods (projected).

• A discussion of ethical considerations.

• A discussion about the methodological limitations : reliability, validity, credibility of research (both internal and external), robustness, transferability and generalizability.

• The mention of some research benefits, merits, and/or possible outcomes.

• Your list of references: only include references that you cite in-text (include a minimum of 12 references, including a min. of 10 peer-reviewed sources).

Proposal: guidelines, criteria, structure

Criteria for evaluating your proposals

See Rubric and Directives -

On CANVAS

Criteria

You can submit an outline (1p)

Outline

What is your gap?

What is your specific question? 1 sentence

Bullet points:

Your purpose?

Your approach?

Your methods (data collection and analysis)

... Be mindful of how you will be able to access the respondents or data you need to conduct the analysis.

the overall logic of the project is paramount

The methods fit the design, which fits the purpose, which fits with the gap...

Criteria<

Suggested structure

Your projects will be 6-8 pages

(double spaced) plus references

Intro (problem statement) 0.75 p.

Literature review: 3-4 pages

Research question and aim (+ expected contribution statement): 3 sentences

Methodology: 2-3 pages.

Discussion and conclusion: 0.5 page

Minimum 12 sources 10 peer reviewed.

: Use sources in the methodology and other sections (discussions, intro, conclusion) too.

Suggested structure

Methodology or methods?

Methods vs methodology

"Methodology" in research is synonym to "design" is quali, quanti, mixed-methods

"methods" are, the means of data collection and data analysis The methods could be synonym to the research's "protocol". For example, interviews, focus groups, surveys (open-ended or close-ended questions), content analysis, participant observations, undercover observations, secondary data analysis, etc. are research "methods".

"Intervention" research

Whether you are proposing a program assessment, writing a report that will guide the strategic planning of your agency, analyzing policies, conducting audit, or satisfaction surveys, you are technically using "research methods".

The management of change, within organization requires stakeholder buy-in. Before producing useful information management can use, it is always good to define collectively the parameters of any analysis,

Organizational development and Intervention

What are you managing? What is the purpose? Management of innovation, knowledge, talents etc.

KEEP IN MIND

Important terms:

Research gap

General question (vs) research question (or "specific question")

Unit of analysis and research scope

Dependent/independent variables

Concepts' operationalization

Online resources on CANVAS

to help you write the proposal

Make an appointment :

The graduate project

The MPA 698s is the program's culminative experience.

In the 698s class, you will develop a project individually on a topic of your choice.

During the pre-enrollment workshops, you will be asked to propose a topic and select peer-reviewed references.

This class and the preparatory workshops will build on what you have learned in the MPA 620.

Learn more about creating dynamic, engaging presentations with Prezi