A Focus-Group Revisit of Our Kumar 2022 Food-Choice Replication: All 10 Phase 2 Sub-Themes Recovered, Plus 4 New Frames From Group Dynamics

In 2022, Kumar, Kulkarni and Rathi published “Evolving Food Choices Among the Urban Indian Middle-Class” in Frontiers in Nutrition. The paper’s Phase 2 analysis used semi-structured interviews with 22 urban middle-class Indians and produced 2 primary themes (“Changing socio-cultural environment” and “Changing food environment”) with 10 sub-themes under them — ranging from globalization and urbanization to lifestyle diseases and food as identity.

We previously replicated that interview pass with 22 synthetic personas, recovering 9 of the 10 sub-themes and surfacing 6 additional frames on top of the paper’s set. That study validated the interview protocol.

This study asks a natural follow-up: what happens if we run the same research territory as a moderated group discussion instead of individual interviews, with the same personas?

We ask two questions: (1) does a moderated group discussion with the same personas still arrive at the sub-themes Kumar documented? And (2) does the group format surface anything the individual-interview format did not?

The answer on both: yes. All 10 of Kumar’s Phase 2 sub-themes surfaced in our two focus groups, plus 4 additional frames the group format added on top of the paper’s set.

Note on Kumar’s actual Phase 1. Kumar et al.’s own paper has two phases. Phase 1 was a pair of focus group discussions on a different topic — consumer perceptions of processed foods — and is not what this study replicates. We plan to replicate Kumar’s Phase 1 directly as a separate study. This page is about the focus-group revisit of the Phase 2 food-choice question territory.

We did not have all this confusion earlier. Food was food. Now food also has fashion.

— Fathima Noushad, 48, Tirur (FGD2, Round 2)

The Original Study

Kumar et al.’s Phase 2 used semi-structured interviews with 22 middle-class Indians (13 women, 9 men, ages 40-65, all college-educated) across Mumbai and Kochi. The three open-ended interview questions, verbatim from Table 2 of the paper, were:

Describe your day-to-day eating habits.
Have your eating habits or food choices changed since childhood?
In general, do you see any changes in eating habits and the food environment? If so, what are the reasons for these changes?

Their thematic analysis produced 2 primary themes — Changing socio-cultural environment (globalization & urbanization, long work hours, rising income levels, decline in household cooking) and Changing food environment (food diversity, availability & accessibility, convenience, advertising & media, food as identity, lifestyle diseases & quality of food) — 10 sub-themes in total. (International fast food chains appear as illustrative content inside Food diversity, not as a separately named sub-theme.)

Paper: Frontiers in Nutrition, PMC9001910

Our Focus-Group Follow-Up

Two moderated focus group sessions using the same personas from our earlier interview replication. Because both studies now use the same synthetic people, any thematic difference between them is attributable to method alone, not to panel composition.

	FGD1 Maharashtra/Central	FGD2 Kerala
Personas	9 (2 Mumbai + 7 Pune)	7 (Ernakulam, Thrissur, Malappuram)
Discussion rounds	3	3
Persona responses	27	21
Source panel	Our 22-persona interview-replication panel	Our 22-persona interview-replication panel

Panel regional breakdown mirrors our interview replication — the 2-Mumbai + 7-Pune composition of FGD1 reflects that our panel has only 2 Mumbai-resident personas; the Kerala panel broadens beyond Kochi to Ernakulam + Thrissur + Malappuram. Both broadenings are deviations documented below.

How the discussion was conducted

Each focus group began with a single opening question to all participants. After each round, the moderator selected the most substantive quotes, summarized each participant’s contribution, named the themes that had surfaced, and chose what to ask next — probing a thread that seemed rich or pivoting to new territory. Participants in the next round saw the moderator’s curated digest of the prior round’s discussion and responded in that context.

Participants were never shown the study objective, a research brief, or the paper’s interview questions. They saw only the scenario and the current question.

The two moderators’ paths, side-by-side

The two focus groups started from the same opening and ended up covering similar territory, but the moderators made meaningfully different choices about sequencing:

	Kumar’s interview anchors	FGD1 moderator’s path (Maharashtra)	FGD2 moderator’s path (Kerala)
Round 1 opening	Describe your day-to-day eating habits.	”Think about what you ate and drank yesterday, from morning to night — can you walk us through it, including where you got the food and who you ate with?"	"To get us started, could you walk us through what you ate and drank yesterday, from morning to night, and who you were with at each point?”
Round 2	Changes since childhood.	”A few of you described weekdays as very controlled — tiffins from home, leftovers, ‘no experiments’… Can you tell me about the rules or habits behind that in your household: who decides what is a normal weekday meal, when outside food is acceptable, and how those habits have changed from a few years ago?” (probed household control)	“A few of you described very regular home food, packed tiffins, and family members joining meals by phone or video. If you compare your meals now with how your household ate 10 or 15 years ago, what feels most different — in the food itself, where it comes from, or who is at the table?” (pivoted to childhood/temporal change — Kumar’s Q2)
Round 3	Food environment and its reasons.	”A lot of you described a very clear line between weekday ‘system food’ and the meals that feel worth looking forward to. Can you tell me about a recent meal that felt special, indulgent, or celebratory…” (advanced to celebratory eating)	“A few of you said the backbone is still home food, but children now ask for things like pasta, shawarma, oats, or food they see online… Can you tell me about one recent meal or food request that felt new in your house — what it was, who wanted it, where it came from, and how the rest of the family reacted?” (probed children + media — reached Kumar’s Advertising & media sub-theme)

That both moderators still arrived in broadly the same thematic territory by distinct routes is itself a signal: the topic was dense enough that multiple conversational paths converge on the same ground.

Sub-Themes Recovered (Both FGDs combined)

Comparing against Kumar et al.’s 10 Phase 2 sub-themes:

#	Paper’s Sub-Theme	Recovered?	Strongest Evidence
1	Globalization & Urbanization	✓ Full	Migration to Pune, Mumbai, Dublin, Manchester, Gulf — husbands and children named explicitly across both panels
2	Long work days & sedentary lifestyles	✓ Full	”Weekday means no experiments” (Rajendra, FGD1); schedule-driven tiffin logic from Meghana, Sudha, Vandana
3	Rising income levels	~ Partial	Now expressed as inflation pressure — “Now I look at one pizza bill and mentally calculate one week of vegetables” (Rajendra, FGD1)
4	Decline in household cooking	✓ Full	Reframed: cooking hasn’t declined, but has been systematized — “one cooking, two boxes” (Meghana), “I cook for management also” (Lekha, FGD2)
5	Food diversity (incl. international fast-food chains as paper’s example)	✓ Full	Pasta, shawarma, oats, cold coffee — all named, mostly as children’s requests. International chains (KFC, Domino’s) appear as parental observations of children’s requests rather than as direct consumption
6	Availability & accessibility	✓ Full	Swiggy, Zomato, Zepto, BigBasket, D-Mart, Lulu hypermarket, local kiranas — explicitly named across both panels
7	Convenience	✓ Full	Quick-commerce mentioned with ambivalence — used but guarded against as temptation
8	Advertising & media	✓ Full	”They also ask for pasta one day, shawarma another day, oats if some YouTube person has recommended it.” (Fathima, FGD2). The session’s theme synthesis explicitly identified “YouTube, Instagram, WhatsApp as active channels bringing new foods into middle-class homes”
9	Food as identity	✓ Full	Regional identity surfaced strongly — kosha mangsho, appam-stew, Tamil keerai, Kerala mathi, cherupayar, tomato rice with cucumber raita
10	Lifestyle diseases & quality of food	✓ Full	BP, diabetes, thyroid, acidity, glucose — “we eat like people who want functioning pancreases at 75” (Nalini, FGD1)

10 of 10 sub-themes recovered. Notably, Sub-theme #8 (advertising and media) did not surface in our interview replication. The focus-group pass brought it in: the FGD2 moderator, responding to what Kerala participants were saying about their children, chose to probe new food requests specifically, and YouTube, Instagram, and WhatsApp came up as the channel.

Four Frames That Emerged On Top of Kumar’s 10 Sub-Themes

These are four frames that surfaced in our group discussions and extend the thematic set beyond Kumar’s original 10. They tend to require multi-round discussion to appear, which is why focus-group formats are better suited to producing them than structured interview protocols.

A fifth observation at the end of this section is a 2026 update on Kumar’s rising-income theme rather than something new.

1. The efficiency-vs-meaning tension

Across FGD1, participants converged on a frame that came to be articulated as: specialness = suspension of optimization. Weekday food is ruthlessly systematized — tiffin logic, leftovers, cost control. What makes a meal “special” is not richer ingredients, but the moment being freed from the weekday management system.

“Remote work changed the whole architecture. Now if I eat outside, it is a deliberate act, not background leakage.” — Sohini Chakraborty, 38, Pune (FGD1, Round 2)

“I do insist on at least one proper sit-down dinner whenever possible. That is a rule I have become more — not less — firm about over time. The rest of the day can be fragmented: tea standing up, lunch from a dabba, fruit between tasks. But one meal should remind us that we are a household, not just two women managing parallel schedules under one roof.” — Kavitha Subramaniam, 48, Pune (FGD1, Round 2)

2. “Controlled absorption” — how new foods actually enter the household

FGD2 surfaced a specific sociological mechanism: new foods don’t arrive by participant adoption. They arrive through children (hostels, school peers, YouTube), pass through mothers who test and adapt them, get mocked but not vetoed by fathers, and settle into the household as experiments rather than replacements.

“They still eat puttu, kadala, rice, fish curry. But they also ask for pasta one day, shawarma another day, oats if some YouTube person has recommended it, cold coffee during exam tension. We did not have all this confusion earlier. Food was food. Now food also has fashion.” — Fathima Noushad, 48, Tirur (FGD2, Round 2)

“The younger officers can order shawarma, burger, whatever chemical experiment is available, and continue working. My generation still wants a steel tiffin from home. There is some psychological comfort in opening a box packed in your own kitchen. It tells you at least one part of the day is not outsourced.” — Mathew Chacko, 46, Aluva (FGD2, Round 2)

3. Meals shared across distance, not across the table

FGD2 returned again and again to the fact that today’s Kerala middle-class meal often includes family members who are not in the room — husbands in the Gulf, children in hostels, daughters in Manchester — present through speakerphone or video. This wasn’t a side comment; it emerged as one of the session’s dominant organizing frames.

“My son on speaker saying hostel canteen food is bad, my daughter on video in Manchester eating toast after shift — these are part of our meals now. We have adjusted. Everyone has adjusted. Still, when I serve fish curry, some part of my mind is counting people who are not in the room.” — Sr. Annamma Pulimoottil, 53, Thrissur (FGD2, Round 2)

4. Invisible women’s labor as managed infrastructure

Both panels surfaced this, and both panels did so most clearly in Round 2 — after participants had heard each other’s Round 1 meal descriptions and the moderator’s curated quotes. None of the men brought it up spontaneously in Round 1. By Round 2, it was explicit.

“One more thing: the rest of the day can be fragmented: tea standing up, lunch from a dabba, fruit between tasks. But one meal should remind us that we are a household, not just two women managing parallel schedules under one roof.” — Kavitha Subramaniam, 48, Pune (FGD1, Round 2)

“Earlier I cooked for appetite. Now I cook for management also.” — Lekha Pulayan, 47, Thrissur (FGD2, Round 2)

A 2026 update on Kumar’s rising-income sub-theme

Not a new theme, but worth flagging: Kumar’s Rising income levels sub-theme documented middle-class incomes enabling new behaviors. Our 2026 personas rotate the same axis 180°: cost pressure organizes their menus. Same underlying economic reality, a different era’s framing.

“Now I look at one pizza bill and mentally calculate one week of vegetables.” — Rajendra Patil, 47, Pune (FGD1, Round 2)

“Now I do one proper hypermarket buy, then local vegetables, then online for some packed items if needed. Earlier it was more nearby shop, fish from market, whatever was seasonal, whatever was available.” — Naseema Abdulla Thottathil, 41, Irinjalakuda (FGD2, Round 2)

Son ate more generously, obviously, and then complained that we eat like patients. I told him we eat like people who want functioning pancreases at 75.

— Nalini Venkataraman, 67, Mumbai (FGD1, Round 1)

Focus-Group Pass vs. Interview Pass: What the Method Comparison Reveals

Because the same 16 personas appear in both studies, we can isolate what the focus-group method contributes beyond the interview method.

What the focus-group pass added:

Brought in the advertising & media sub-theme. Kumar’s Advertising & media sub-theme did not surface in our interview pass. The focus-group pass surfaced it directly through discussion of children’s new food requests.
Revealed cross-round stance dynamics. Men’s acknowledgment of women’s invisible labor appeared in Round 2 — after Round 1’s discussion had been digested by the group. It didn’t exist in Round 1.
Produced emergent sociological frames. “Controlled absorption”, “specialness = suspension of optimization”, “meals shared across distance” — these are synthesis frames the group converged on, not individual persona claims.
Made different panel paths visible. Two moderators, same opening, meaningfully different sequencing — information about how the same research space can be navigated multiple ways.

What the focus-group pass did not change:

Core theme-recovery parity with the interview pass (and one theme more).
Persona voice fidelity — specific dishes, brands, family members, locations appear in both.

What the focus-group pass lost vs. the interview pass:

Per-persona narrative depth. Individual interviews produced more biographical detail per person than 3 group rounds can.

The practical implication: the two formats are complements, not substitutes. Run the interview format for breadth and individual depth. Run the focus-group format when you want emergent frames, stance shifts under social reinforcement, and evidence of how a topic evolves in group context.

Methodology

Dimension	Original Study (Phase 2 interviews)	Our Focus-Group Follow-Up
Method	22 semi-structured interviews	2 moderated synthetic focus groups
Participants	22 real people	16 synthetic personas (subset of our 22-persona interview-replication panel)
Group composition	Individual interviews, Mumbai + Kochi	FGD1: 9 western Maharashtra (2 Mumbai + 7 Pune); FGD2: 7 central Kerala
Language	English (all interviews)	English with Malayalam phrasings allowed
Question protocol	Three anchor questions, same for all participants	Single opening question + moderator-adapted follow-ups across three rounds
Cross-pollination	None (individual format)	Moderator-curated quotes and per-persona summaries from the previous round, shown to each participant before their next turn
Fieldwork date	Not stated in paper (manuscript received Dec 2021)	April 2026
Time to results	Weeks to months	Minutes

The moderator-curated multi-round format is not a perfect replica of real focus-group turn-taking — participants don’t hear every word others say, only the moderator’s curated digest. But it preserves the core property that matters: each persona’s next response is shaped by what the group has already said, mediated by a moderator who also chooses what to ask next. That’s enough for emergent group frames to appear.

Deviations From the Original Protocol

#	Deviation	Reason	Impact
1	Follow-up questions generated adaptively by the moderator, not asked verbatim from the paper’s anchor list	A moderated focus-group format requires giving the moderator authority to adapt — hand-specifying every question removes the point of running a focus group.	Both moderators arrived at Kumar’s Q2 and Q3 thematic territory by their own navigation.
2	FGD1 panel is Pune-dominant western Maharashtra (2 Mumbai + 7 Pune)	Our interview-replication panel has only 2 Mumbai-resident personas. Reusing that panel gives direct within-persona comparability — a methodological gain that outweighs strict city matching.	Pune vs. Mumbai food cultures share a Maharashtrian core but differ in cost-of-living and cosmopolitan exposure.
3	FGD2 broadens from Kochi to central Kerala (Ernakulam, Thrissur, Malappuram)	Same panel-reuse rationale as #2.	Regional diversity is a feature — surfaces Thrissur and Malappuram angles a Kochi-only panel would not.
4	No real cross-talk between participants	Synthetic focus groups mediate cross-pollination through moderator-curated digests rather than direct speech.	Closer to a skilled-moderator focus group than to spontaneous group dynamics.
5	Same 16 personas as our interview replication (within-subjects method comparison)	Panel reuse isolates method from panel composition.	Any focus-group vs. interview thematic difference is attributable to method alone.

Limits

No real participants as ground truth. This study compares to a published paper, not fresh interviews with the same personas’ real-world equivalents.
Two focus groups at 9 and 7 personas — a convenience split of our 22-persona interview-replication panel. Scaling would narrow uncertainty on emergent frames but dilute the method-comparison claim.
The temporal gap between the paper and this run. Kumar et al.’s fieldwork year isn’t stated in the paper (manuscript received December 2021); ours is April 2026. Some differences — cost pressure instead of rising-income optimism, YouTube/Instagram as dominant channels alongside TV — reflect real change in the food environment, not method divergence.

What This Means

For researchers: a moderated multi-round synthetic focus group with the same personas used in an interview replication recovered all 10 of a published study’s documented Phase 2 sub-themes on the same demographic — including one that had not surfaced in the interview format. The group format produces content that fixed-question formats structurally do not: stance shifts across rounds, emergent interpretive frames, and the social mechanisms by which change enters a stable cultural system.

For practitioners: if you want the sub-themes a structured three-question pass can produce, the interview format is faster and equally reliable. If you want to see how a topic evolves under social reinforcement — which is what most product, brand, and policy questions actually require — the focus-group format produces information the interview format cannot.

If you want to test this on a question you care about, book a demo or explore more evidence.