Skip to main content

Table 2 Summary of published/unpublished work on experiential household food security scales, India, 2000–2015

From: Internal validity and reliability of experience-based household food insecurity scales in Indian settings

ID Authors Setting Period of study Sample size Study population Respondent Language Type of scale Recall period Internal reliability/validity Food security status
1 Nord et al. [11]a Rural Orissa 2000–2001 282 All adults and children in the household Oriya 9-item scale adapted from 18-item US HFSSMb 30 days In-fit: 0.75–1.11
High out-fit: balanced meal, ate less and cut-skip meal
Food secure: 8.0%
Food insecure: 92.0% (with 57.0% food insecure with hunger)
2 MSF [23] Chittagong Hill Tracts Sajek Union bordering Bangladesh and Mizoram, India March–April 2008 151 households from 34 villages Children aged 1–5 years Adult females/males in the household. Bengali 9-item FANTA HFIAS version 3 30 days Food secure: 4.0%
Mildly food insecure: 4.6%
Moderately food insecure: 11.9%
Severely food insecure: 79.5%
3 Agarwal et al. [21] Slum in north-east Delhi June–July 2008 410 All adults in household Adult female involved in cooking/purchasing food Hindi 4-item scale adapted from 6-item shorter version of US HFSSMc 12 months Cronbach’s α: 0.8
In-fit: 0.77–1.07
High out-fit: cut meal/skip meal
Food secure: 9.0%
Food insecure: 51% (with 27.1% food insecure without hunger and
23.9% food insecure with hunger)
4 Agarwal et al. [22] 75 slums of Meerut city October 2007–March 2008 40,016 Women of reproductive age (WRA) Adult female Hindi 4-item scale adapted from 6-item shorter version of US HFSSMd 12 months Point bi-serial correlation: 0.43–0.59
Cronbach’s α: 0.725
In-fit - 0.52–1.11
High out-fit: nutritious food
Food insecure: 74.2% (with 31.5% food insecure without hunger and 42.7% with hunger)
5 Pasricha et al. [24] Rural Karnataka 415 Children aged 12–23 months Mothers of children Kannada 9-item FANTA HFIAS version 3 30 days   Food secure: 42.0%
Food insecure: 58.0%
6 Gopichandran et al. [14] Tamil Nadu, Vellore (urban) May/June 2009 130 All adults and children in household Head of household/housewife Tamil 18-item USHFSSM 12 months Food secure: 25.4%
Food insecure with hunger: 61.5%
Food insecure without hunger: 13.1%
7 Mukhopadhyay et al. [18] Bankura-1 CD block district, West Bengal July–August 2009 267 tribal households All adults in household Head of household/responsible adult family member, preferably a woman Bengali A validated Bengali version of the US 6-item short form food security scale 12 months κ: > 0.84
Cronbach’s α: 0.82
47.2% households had high or marginal food security, whereas 29.6% and 23.2% had low and very low food security, respectively
The prevalence of low and very low food security was higher among households having children aged under-five (31.2 and 24.3%, respectively) compared to households without children under-five (24.6 and 20.0%, respectively)
8 Mukhopadhyay and Biswas [19] Bankura-1 CD block district, West Bengal July–August 2009 188 tribal households Tribal children aged 24–59 months Mothers of the children Bengali A validated Bengali version of the 6-item shorter version of US HFSSM 12 months Same as above High/marginal food security: 46.8%
Low food security: 28.7% Very low food security: 26.5%
9 Chatterjee et al. [25] Three slums in north-west Mumbai January–March 2010 283 All adult members Adult female Marathi 9-item FANTA
HFIAS
30 days Food secure: 23.7%
Severely food insecure: 59.7%
Mild to moderate food insecure: 16.6%
10 Chinnakali et al. [26] A resettlement colony in south Delhi   250 All adult members Females aged 18–50 years Hindi 9-item FANTA HFIAS 30 days Food insecure: 77.2% (with 49.2% households being mildly food insecure, 18.8% households moderately food insecure and 9.2% households severely food insecure)
11 Maitra [30] 15 slums in Kolkata Metropolitan Corporation (KMC) area April 2010–January 2011 500 All adults and children Head of household/adult female in charge of kitchen Bengali 9-item adult scale and 5-item child scale adapted from 18-item US HFSSMe 30 days 9-item adult scale
Cronbach’s α: 0.85
Rasch reliability: 0.75
Sensitivity: 0.83
Specificity: 0.97
Positive predictive value: 0.85
In-fit: 0.7–1.3
High out-fit: never cooked rich meal
5-item child scale
Rasch reliability: 0.94
Sensitivity: 0.87
Specificity: 0.97
Positive predictive value: 0.89
In-fit: 0.63–1.25
High out-fit: child could not eat three square meals
9-item adulteverscale
Food insecure: 15.4% (with 12.8% moderately food insecure and 2.6% severely food insecure)
Food secure: 84.6% (with 76.2% highly food secure and 8.4% marginally food secure)
5-item child scale
Food insecure: 20.4% households
Highly food secure: 70.7% households (with 7.9% households having children who were marginally food secure)
12 UHRC [15] Slum in north-east Delhi June–July 2011 232 Children aged below 5 years Adult female aged ≥18 years involved in cooking and purchasing food Hindi 8-item child food security scale based on US HFSSM 30 days   Households where children are food secure: 0.4%
Marginally food secure: 8.6%
Food insecure without hunger: 27.6%
Food insecure with hunger: 63.4%
13 Gupta et al. [16] Four Delhi slums August 2011–October 2012 446 WRA (15–45 years) with children (6–35 months) Mothers involved in cooking and purchasing food Hindi 8-item child scale based on 18-item USHFSSM 12 months   Food secure: range 45.0–81.0%
Low food insecurity: range 18.0–49.0%
Very low food insecurity: range 1.0–15.0%
14 Gupta et al. [17] Four Delhi slums 2012 446 WRA (15–45 years) with children (6–36 months) Mothers involved in cooking and purchasing food Hindi 8-item child scale based on 18-item USHFSSM 12 months   Food insecure: 38.1% of children
15 Nord and Cafiero [29] Rural/urban India 2012 Gallup World Poll Survey 2540 Adult men and women aged 15+ years, children aged <15 years Randomized adult (aged 15 +) in household Multiple. Hindi (n = 1480), Marathi (n = 280), Bengali (n = 230), Telegu (n = 210), smaller numbers of Gujarati, Kannada, Malayalam, Tamil, Oriya, Punjabi and Assamese FIESf
8 adult items
7 child items
12 months Adult scale
In-fits: 0.7–1.3 except item ‘ran out’ (1.38)
Rasch reliability: 0.72
Child scale (6 items)
In-fits: acceptable range 0.7–1.3
Combined adultchild scale
In-fits: acceptable range 0.7–1.3
Not available
16 IIPS-UNICEF [28] Maharashtra state —aggregate/rural/urban February–April 2012 2630 Household with children below 2 years of age Household member, primarily involved in the food preparation Marathi/English FANTA 9-item HFIAS 30 days Maharashtra basic dichotomous (yes/no) scale
Cronbach’s α: 0.91
Rasch reliability: 0.818
In-fits: 0.62–1.29
High out-fits: worried (4.50), preferred food (5.99), hungry (7.04) and whole day (11.54)
Maharashtra polytomous scale
Rasch reliability: 0.807
In-fits: 0.77–1.45
In-fit for worried is high
High out-fit: worried, hungry and whole day
Food secure: 57.0%
Mildly food secure: 17.0%
Moderately food secure: 13.0%
Severely food insecure: 14.0%
17 Wright and Gupta [20] A slum from north-east Delhi 2010 105 Convenience sample of members in families receiving care at health centres and clinics WRA (18–45 years) who were responsible for food procurement Hindi 6-item shorter version of US HFSSM 12 months   Food insecure: 57.0%
18 Nord and Cafiero [29] Rural/urban India 2014 Gallup World Poll Survey 3000 (nationally representative sample) Men and women aged 15+ years Randomized adult (aged 15+ years) in household Multiple. Hindi (n = 1480), Marathi (n = 280), Bengali (n = 230), Telegu (n = 210), smaller numbers of Gujarati, Kannada, Malayalam, Tamil, Oriya, Punjabi and Assamese Extended FIES 8 adult items (plus 2 child items if child aged under-five lived in the household)g 12 months In-fit: 0.7–1.3 for 8 items and excellent for 7 (0.8–1.2)
Rasch reliability: 0.72
Extended FIES (hungry and whole day trichotomous)
All Rasch-Thurstone in-fit statistics were in an excellent range 0.8–1.2
Overall item in-fit statistics good for all 8 items (0.7–1.3)
Rasch reliability: 0.82
Not yet released
19 IFPRI [27] Odisha 2015 9-item FANTA HFIAS 30 days In-fits: 0.92–1.36 (high in-fit—preferred food)
High out-fit: preferred food
Not available
  1. All studies are household-based cross-sectional studies
  2. aThe survey and initial adaptation of the 18-item US HFSSM in rural Odisha was undertaken by Nikhil Raj and Anup Satpathy. Dr. Mark Nord joined the project at a later stage to undertake the psychometric assessment of the scale. The original paper is cited as: Raj and Satpathy [34]
  3. bThe items ‘adult/child cut size-skip meal’ have the follow-up question ‘how often’
  4. c,dThree items have the follow-up question ‘how often’. Only ‘slept hungry’ had no follow-up
  5. eAll items except ‘did you get rich food’ and ‘did adult/child lose weight’ were followed by frequency of occurrence
  6. fNo follow-up question ‘how often’
  7. gExtension was to include ‘how often’ follow-up questions to the two most severe questions (hungry and whole day). Response options: ‘only once or twice’, ‘in some months but not every month’, ‘almost every month’. Partial credit Rasch model analysis. Child items were for research purposes and not included in the scale