Session 2 Exposition - "What are constructs in MKTR?"

Hi all,

I got feedback from many of you saying the concept of "constructs" in MKTR is unclear.

I hope this blog post explaining the same, with some illustrations and explanations clarifies things.

Definition:

"Constructs are mental objects linked to behavior." (for our purposes in MKTR).

The behavior may be "expressed" or may be "latent" (i.e. yet to manifest or exist only as a propensity).

The confusion, I think, stems from the fact that sometimes we label the behaviors themselves as constructs (e.g. Cash on delivery), and sometimes the mental objects driving the behavior as a construct.

For clarity, we shall now on use "construct" only for the mental object that manifests as behavior.

Nevertheless, we maintain that under the 'right' conditions - the intersection of motive, means and opportunity - constructs tend to manifest as expressed behavior.

Some Examples:

Consider some simple examples: "Brand Preference" and "Purchase intention" are both constructs in that they are mental objects.

They may manifest as behavior and when they do they will have measurable business outcomes.

This begs the Q, how does one even start to measure a 'mental object'?

As stated in class y'day, we try to do so in two ways - by measuring the motivations behind the behavior and by measuring what we can of the components or features of the behavior itself (we'll get some greater clarity on this later part in Session 4: Qualitative research).

In class, I gave a simple example wherein I tried deconstructing "Purchase Intention" into constituent one-dimensional aspects (see below).

Of course, I pulled out the 2 aspects out of thin-air as it was merely for demonstration purposes. In real life, given the context-sensitive nature of constructs, they are broken down only after extensive exploratory/qualitative work has been undertaken.

Similarly, complex concepts such as "consumer attitude" (towards, say tobacco/ cosmetics/ iProducts etc) or "brand image" are also constructs in that they are mental objects and can be linked to expressed behavior under the right conditions.

Relationship with "Consumer Needs":

For convenience, one can think of constructs as an intermediary step between consumer need and consumer behavior.

For instance, "Purchase intention" is not really a 'consumer need' but merely the means to satisfying one. And "Purchase intention" is a mental object that is not yet expressed behavior.

Hence, it is an easy and clear case of being a "construct".

Take "brand image". What needs does this construct serve?

It depends on the product category etc but one can, on average, think of needs such as mitigation of (quality) risk, search costs, routine/intertia, need to belong/conformity, aspirational reach etc that might be in play.

Two Broad Measurement Approaches:

One is the conventional way we went about in class - We take a behavior pattern of interest (either expressed or latent) and postulate the existence of constructs that are linked to the behavior.

We then break-down this constructs into Likert-able aspects and collect data on the same.

After data collection, we can confirm whether the aspects indeed add up to the constructs as we had hypothesized (e.g., via confirmatory factor analysis, correspondence analysis or structural equation modeling, etc.)

Some of these analyses I mentioned we'll see later in the course.

The above is the approach taken in this complex example of real research on brand advocacy I presented in class (see slide below):

The second approach, which we will see in session 3 - involves collecting bits and pieces of known behavior patterns and then exploring whether there exists a "common", correlation-based inter-relationship among the behaviors (or tendencies) of interest. If yes, then perhaps some construct(s) underlie(s) them.

Class example with the quick-check:

I'm yet to go over the responses to the first quick-check (see slide below) but let me first pin my thoughts down about it.

Thought it best to tabulate my thoughts. Pls click on the image below for a larger view.

Again, I'm not claiming my musings are "right" as opposed to "wrong" etc. Its just a single, simple case of applying analytic thinking and frameworks we picked up in session 2 to a real world, real-time Business problem.

And for this course to be relevant, we have to keep darting between real-life problems - both offline and web-based. I continue to hope for newer insights and perspectives from you folks, who are closer to the consumer-tech revolution as it is happening than an ancient like moi.

Well, that's it for now. Pls do share your thoughts and comments (and even more welcomingly, your own examples, links and ideas) in the comments section below.

Sudhir

Session 2 Updates

Hi all,

Session 2 got done today. It was a loaded one - particularly the introduction to what we mean by constructs in the MKTR context.

We ventured into psychometric scaling and attempted to measure complex constructs using the Likert scale, among others. We also embarked on a common-sensical approach to survey design.

One issue with a lot of MKTR is that it is context-sensitive, so its hard to proclaim 'right' answers that will hold true in general. "It depends" is usually a better bet. Wherever possible I do try to point out general principles and frameworks but in many cases, the problem context decides whether something is true or not.

Well, I will put up a blog post on the subject of "Constructs in MKTR" in the next couple of days. Watch this space.

There were some issues also with the creat.st software but overall, I thought things went by fairly smoothly.

The plan is to share both representative and outlying answers from all sections up here on the blog for enhanced perspective.

I'm also working on a cloud based version of R's software that can allow folks to load data and run R directly off the browser without worrying about the backend code etc. Let's see how that goes.

The homeworks for session 2 are up (see post below).

One other things: Pls form MKTR project groups of upto *5* people each. Previously, I'd said 'exactly 5' people.

Sudhir

Session 2 Homework

Hi all,

For your Session 2 homework, pls fillup the following two surveys.

Big 5 Personality Factors Survey

Values and Life Styles Survey

The surveys may each take upto 20 minutes of your time. There are no 'right' or 'wrong' answers, its just each person's own subjective perception.

Timely and complete submissions alone carry points.

Deadline: is Sunday (07-sept) midnight.

Any queries or concerns, contact me.

Sudhir

Group Formation Instructions

Hi all,

Looking at the number of registrations we have, pls form MKTR project groups of upto *5* people each. Previously, I'd said 'exactly 5' people.

Across section-groups are fine.

Appoint a 'group rep' to manage all group communication with the AAs and me.

Pls select as group name any good quality brand you are aware of. Less common names would be preferred. In case two groups choose the same name, the 'first come first served' rule applies.

Pls find below a google form. The group rep can fill it up giving details of group name and group members' names and PGIDs.

Google form to fill up group details

Fill the form only *once* per group.

Group formation Deadline: By the end of session 4. After that we'll allocate the remaining folks into groups.

After group formation, we'll release the groups list and you can check and confirm that there are no discrpancies.

Any queries etc, contact the AAs directly.

Sudhir

Session 1 Homework

Hi all,

Homework for session 1:

Pls watch this ~ 20 minute video carefully. It features Scott McDonald of Condé Nast holding fort on where MKTR is headed.

“Social Technological and Economic forces affecting Marketing Research over the next decade”

Now, for your HW, pls answer a few simple Qs (True-False, fill in the blanks variety) about the above talk in the following survey:

Questions for Session 1 Homework.

HW Notes:

(i) This is an individual-only HW. Since it involves no R, consulting peers is not permitted.

(ii) I found that using earphones works great in making out what the speaker is saying much more clearly than ordinary speakers. FYI.

(iii) Deadline: The HW should be completed and submitted latest by midnight 8-September.

Any Qs etc, pls feel free to email me or use the comments section below.

Sudhir Voleti

Welcome Message and Session 1 Updates

Hi Co2015,

The list of final registrants is out. Welcome to MKTR.

The first session got done on Tuesday. We covered some subject preliminaries and the crucial task of problem formulation. I've put up the session slides on LMS in PDF form.

About this blog:

This is an informal blog that concerns itself solely with MKTR affairs. It's informal in the sense that its not part of the LMS system and the language used here is more on the casual side. Otherwise, its a relevant place to visit if you're taking MKTR. Pls expect MKTR course related announcements, R code, Q&A, feedback etc. on here.

Since last year, I've made this blog the single-point contact for all relevant MKTR course related info. LMS will only be used for file transfers and email notifications to the class only in emergencies. Each session's blog-post will be updated with later news coming at the top of the post. Kindly bookmark this blog and visit regularly for updates.

About Session 2:

Session 2 deals with psychographic scaling techniques and delves into the intricacies of defining and measuring "constructs" - complex mental patterns that manifest as patterns of behavior. This session also sets the stage for questionnaire design to make an entry in Session 3.

There will be a series of 2 homework assignments for session 2. These concern merely filling up surveys that I will send you (this data we will use later in the course for perceptual mapping and segmentation). Nothing too troublesome, as you can see.

Pre-reads for Session 2:

Pls read the HBS note "Marketing Research" (serial no. 3 in the coursepack) as pre-read for Session 2.

Installing R and RStudio.

The .exe files for R and Rstudio are already placed in LMS in a separate folder. Download them.

Install R first by clicking on its .exe and following instructions.

Next install RStudio by clicking on it and following instructions.

We'll need RStudio up and running by session 3, when we'll have our first brush with R in the classroom.

Any Qs etc, pls feel free to email me or use the comments section below.

Sudhir Voleti

Helping out a Military Man

Was at the CDM (College of Defence Management) in Secunderabad a couple of weeks ago to give a talk to the student officers there on primary data collection methods.

The talk went rather well. Here's an email I received from one of the student officers there. I've anonymized the name to "PS" but the rest is as is.

Sir,

First of all let me congratulate you for educating us on primary data collection methods at the College of defence Mgt, Secunderabad on 26 Jul 14. As a participant offr, undergoing the Higher Defence Mgt Course at the College, I must admit that lots of doubts lingering in our minds on constructs and questionnaire design got clarified after the talk. All the participants were very enthusiastic which was evident from the huge applause and brisk clapping at the end of the talk. Thank you very much.

Sir, my dissertation topic is ‘Space Warfare and Military Strategy’ and the draft hypothesis on which I am working on is “Weaponisation of space in the south Asian context is inevitable”.

Sir, in your pre-lunch interaction, you had mentioned about certain tools which are available on open source for hypothesis testing on such subjects. While I am in the process of preparing a questionnaire for circulating it amongst the participant officers, I would request you to kindly guide me on the issue.

PS

The following is my response.

Hi PS,

Pls call me Sudhir. I am happy for the opportunity to give talks at the CDM and interact with folks like you.

Re your thesis topic, some of my first impressions:

>> Sir, my dissertation topic is ‘Space Warfare and Military Strategy’ and the draft hypothesis on which I am working on is “Weaponisation of space in the south Asian context is inevitable”.

Before drafting hypotheses, its critically important to define the phenomena of interest (upon which your constructs will eventually be based). What do we mean by 'space warfare'? By 'weaponization'? Even by 'south asia' given that space is a, well, global game and we're looking at it with regional lenses?

Once you have built your definitions (and these could be with the help of (i) existing literature, (ii) expert commentary or agreement and/or (iii) context specific understanding), you can start generating ideas for plausible hypotheses.

E.g., define thresholds - level 1 of space militarization is surveillance capability deployment of upto 1 m resolution on target countries, level 2 is so-and-so and so on. Pretty much like how in the IAF you have third, fourth and fifth gen planes and so on based on certain criteria.

By defining levels, generations or other thresholds, you are able to use these constructs directly in your hypothesis construction.

By the way, all this is illustrative only. I have no idea about space weaponization at all.

>> Sir, in your pre-lunch interaction, you had mentioned about certain tools which are available on open source for hypothesis testing on such subjects.

Well, I was referring to text analytic tools that are readily available on open source R statistical platform. So if you have a large body of written material - journal articles, research reports, press reports, expert interview transcripts etc on the subject of your study, you can condense the text to extract meaningful themes from it.

I intend to conduct a workshop with interested CDM faculty where I will share methods and code for text analytics (among other things) with the faculty there. My understanding is that they in turn will pass on the how-to of text analytics to you.

Hope that helps,

Cheers,

Sudhir