For detailed information, visit the Tucson Equity Priority Index StoryMap.Download the layer's data dictionaryWhat is the Tucson Equity Priority Index (TEPI)?The Tucson Equity Priority Index (TEPI) is a tool that describes the distribution of socially vulnerable demographics. It categorizes the dataset into 5 classes that represent the differing prioritization needs based on the presence of social vulnerability: Low (0-20), Low-Moderate (20-40), Moderate (40-60), Moderate-High (60-80) High (80-100). Each class represents 20% of the dataset’s features in order of their values. The features within the Low (0-20) classification represent the areas that, when compared to all other locations in the study area, have the lowest need for prioritization, as they tend to have less socially vulnerable demographics. The features that fall into the High (80-100) classification represent the 20% of locations in the dataset that have the greatest need for prioritization, as they tend to have the highest proportions of socially vulnerable demographics. How is social vulnerability measured?The Tucson Equity Priority Index (TEPI) examines the proportion of vulnerability per feature using 11 demographic indicators:Income Below Poverty: Households with income at or below the federal poverty level (FPL), which in 2023 was $14,500 for an individual and $30,000 for a family of fourUnemployment: Measured as the percentage of unemployed persons in the civilian labor forceHousing Cost Burdened: Homeowners who spend more than 30% of their income on housing expenses, including mortgage, maintenance, and taxesRenter Cost Burdened: Renters who spend more than 30% of their income on rentNo Health Insurance: Those without private health insurance, Medicare, Medicaid, or any other plan or programNo Vehicle Access: Households without automobile, van, or truck accessHigh School Education or Less: Those highest level of educational attainment is a High School diploma, equivalency, or lessLimited English Ability: Those whose ability to speak English is "Less Than Well."People of Color: Those who identify as anything other than Non-Hispanic White Disability: Households with one or more physical or cognitive disabilities Age: Groups that tend to have higher levels of vulnerability, including children (those below 18), and seniors (those 65 and older)An overall percentile value is calculated for each feature based on the total proportion of the above indicators in each area. How are the variables combined?These indicators are divided into two main categories that we call Thematic Indices: Economic and Personal Characteristics. The two thematic indices are further divided into five sub-indices called Tier-2 Sub-Indices. Each Tier-2 Sub-Index contains 2-3 indicators. Indicators are the datasets used to measure vulnerability within each sub-index. The variables for each feature are re-scaled using the percentile normalization method, which converts them to the same scale using values between 0 to 100. The variables are then combined first into each of the five Tier-2 Sub-Indices, then the Thematic Indices, then the overall TEPI using the mean aggregation method and equal weighting. The resulting dataset is then divided into the five classes, where:High Vulnerability (80-100%): Representing the top classification, this category includes the highest 20% of regions that are the most socially vulnerable. These areas require the most focused attention. Moderate-High Vulnerability (60-80%): This upper-middle classification includes areas with higher levels of vulnerability compared to the median. While not the highest, these areas are more vulnerable than a majority of the dataset and should be considered for targeted interventions. Moderate Vulnerability (40-60%): Representing the middle or median quintile, this category includes areas of average vulnerability. These areas may show a balanced mix of high and low vulnerability. Detailed examination of specific indicators is recommended to understand the nuanced needs of these areas. Low-Moderate Vulnerability (20-40%): Falling into the lower-middle classification, this range includes areas that are less vulnerable than most but may still exhibit certain vulnerable characteristics. These areas typically have a mix of lower and higher indicators, with the lower values predominating. Low Vulnerability (0-20%): This category represents the bottom classification, encompassing the lowest 20% of data points. Areas in this range are the least vulnerable, making them the most resilient compared to all other features in the dataset.

For detailed information, visit the Tucson Equity Priority Index StoryMap.Download the Data DictionaryWhat is the Tucson Equity Priority Index (TEPI)?The Tucson Equity Priority Index (TEPI) is a tool that describes the distribution of socially vulnerable demographics. It categorizes the dataset into 5 classes that represent the differing prioritization needs based on the presence of social vulnerability: Low (0-20), Low-Moderate (20-40), Moderate (40-60), Moderate-High (60-80) High (80-100). Each class represents 20% of the dataset’s features in order of their values. The features within the Low (0-20) classification represent the areas that, when compared to all other locations in the study area, have the lowest need for prioritization, as they tend to have less socially vulnerable demographics. The features that fall into the High (80-100) classification represent the 20% of locations in the dataset that have the greatest need for prioritization, as they tend to have the highest proportions of socially vulnerable demographics. How is social vulnerability measured?The Tucson Equity Priority Index (TEPI) examines the proportion of vulnerability per feature using 11 demographic indicators:Income Below Poverty: Households with income at or below the federal poverty level (FPL), which in 2023 was $14,500 for an individual and $30,000 for a family of fourUnemployment: Measured as the percentage of unemployed persons in the civilian labor forceHousing Cost Burdened: Homeowners who spend more than 30% of their income on housing expenses, including mortgage, maintenance, and taxesRenter Cost Burdened: Renters who spend more than 30% of their income on rentNo Health Insurance: Those without private health insurance, Medicare, Medicaid, or any other plan or programNo Vehicle Access: Households without automobile, van, or truck accessHigh School Education or Less: Those highest level of educational attainment is a High School diploma, equivalency, or lessLimited English Ability: Those whose ability to speak English is "Less Than Well."People of Color: Those who identify as anything other than Non-Hispanic White Disability: Households with one or more physical or cognitive disabilities Age: Groups that tend to have higher levels of vulnerability, including children (those below 18), and seniors (those 65 and older)An overall percentile value is calculated for each feature based on the total proportion of the above indicators in each area. How are the variables combined?These indicators are divided into two main categories that we call Thematic Indices: Economic and Personal Characteristics. The two thematic indices are further divided into five sub-indices called Tier-2 Sub-Indices. Each Tier-2 Sub-Index contains 2-3 indicators. Indicators are the datasets used to measure vulnerability within each sub-index. The variables for each feature are re-scaled using the percentile normalization method, which converts them to the same scale using values between 0 to 100. The variables are then combined first into each of the five Tier-2 Sub-Indices, then the Thematic Indices, then the overall TEPI using the mean aggregation method and equal weighting. The resulting dataset is then divided into the five classes, where:High Vulnerability (80-100%): Representing the top classification, this category includes the highest 20% of regions that are the most socially vulnerable. These areas require the most focused attention. Moderate-High Vulnerability (60-80%): This upper-middle classification includes areas with higher levels of vulnerability compared to the median. While not the highest, these areas are more vulnerable than a majority of the dataset and should be considered for targeted interventions. Moderate Vulnerability (40-60%): Representing the middle or median quintile, this category includes areas of average vulnerability. These areas may show a balanced mix of high and low vulnerability. Detailed examination of specific indicators is recommended to understand the nuanced needs of these areas. Low-Moderate Vulnerability (20-40%): Falling into the lower-middle classification, this range includes areas that are less vulnerable than most but may still exhibit certain vulnerable characteristics. These areas typically have a mix of lower and higher indicators, with the lower values predominating. Low Vulnerability (0-20%): This category represents the bottom classification, encompassing the lowest 20% of data points. Areas in this range are the least vulnerable, making them the most resilient compared to all other features in the dataset.

Kazakhstan Ratio: Net Income before Tax to Total Assets (ROA) data was reported at 3.950 NA in Jan 2025. This records a decrease from the previous number of 4.610 NA for Dec 2024. Kazakhstan Ratio: Net Income before Tax to Total Assets (ROA) data is updated monthly, averaging 2.135 NA from Dec 2007 (Median) to Jan 2025, with 206 observations. The data reached an all-time high of 13.240 NA in Nov 2010 and a record low of -24.060 NA in Dec 2009. Kazakhstan Ratio: Net Income before Tax to Total Assets (ROA) data remains active status in CEIC and is reported by Agency of the Republic of Kazakhstan for Regulation and Development of Financial Markets. The data is categorized under Global Database’s Kazakhstan – Table KZ.KB029: Second Tier Banks: Profitability Indicators.

This dataset examines financial inclusion and bank stability in Ethiopia, containing panel data from 17 commercial banks over the period 2015-2023. In 2015, there were 17 commercial banks in Ethiopia but to maintain confidentiality, the names of commercial banks have been anonymized and are referred to by generic labels: 1, 2, 3, 4..., and 17. This process allows the dataset to be analyzed and shared openly in support of reproducibility and transparency in research.VariablesBank Stability (ZS): Computed using the Z-score to measure stability.Financial Inclusion Index (IFI): Developed using two-stage Principal Component Analysis (PCA) with 10 conventional and 5 digital indicators.Loan to Deposit Ratio (LDR): Computed based on the loan to deposit ratio.Provision to Loan (PL): Computes the loan loss provision ratio.Natural Logarithm of Total Assets (lnTA): Logarithmic form of total assets.Capital Adequacy Ratio (CAR): Computed by Tier-1 capital and Tier-2 capital divided by risk-weighted assets.Income Diversification (IND): Computed based on the non-interest income to total income ratio.Operational Efficiency Management (EF): Measured using Data Envelopment Analysis (DEA) with five input variables (salary and benefits, provisions, general expenses, branches, and deposits) and two output variables (net interest income and non-interest income).Real Lending Interest Rate (RLIR): Inflation-adjusted interest rate.GDP Growth Rate (GDP): Annual percentage change in GDP.This dataset provides comprehensive insights into the relationships between financial inclusion and bank stability, supporting future research and policy formulation.

Kazakhstan Ratio: Net Income before Tax to Equity (ROE) data was reported at 27.870 NA in Jan 2025. This records a decrease from the previous number of 32.830 NA for Dec 2024. Kazakhstan Ratio: Net Income before Tax to Equity (ROE) data is updated monthly, averaging 16.140 NA from Dec 2007 (Median) to Jan 2025, with 206 observations. The data reached an all-time high of 2,920.830 NA in Dec 2010 and a record low of -2,737.880 NA in Jan 2010. Kazakhstan Ratio: Net Income before Tax to Equity (ROE) data remains active status in CEIC and is reported by Agency of the Republic of Kazakhstan for Regulation and Development of Financial Markets. The data is categorized under Global Database’s Kazakhstan – Table KZ.KB029: Second Tier Banks: Profitability Indicators.

Abstract

The Integrated Living Conditions Survey (ILCS), conducted annually by the NSS National Statistical Service of the Republic of Armenia, formed the basis for monitoring living conditions in Armenia. The ILCS is a universally recognized best-practice survey for collecting data to inform about the living standards of households. The ILCS comprises comprehensive and valuable data on the welfare of households and separate individuals which gives the NSS an opportunity to provide the public with up to date information on the population’s income, expenditures, the level of poverty and the other changes in living standards on an annual basis.

Geographic coverage

Urban and rural communities

Analysis unit

• Households;
• Individuals.

Kind of data

Sample survey data [ssd]

Sampling procedure

During the 2001-2003 surveys two-stage random sample was used; the first stage covered the selection of settlements - cities and villages, while the second stage was focused on the selection of households in these settlements. The surveys were conducted on the principle of monthly rotation of households by clusters (sample units). In 2002 and 2003 the number of households was 387 with the sample covering 14 cities and 30 villages in 2002 and 17 cities and 20 villages in 2003.

During the 2004-2006 surveys the sampling frame for the ILCS was built using the database of addresses for the 2001 Population Census; the database was developed with the World Bank technical assistance. The database of addresses of all households in Armenia was divided into 48 strata including 12 communities of Yerevan city. The households from other regions (marzes) were grouped according to the following three categories: big towns with 15,000 and more population; villages, and other towns. Big towns formed 16 strata (the only exception was the Vayots Dzor marz where there are no big towns). The villages and other towns formed 10 strata each. According to this division, a random, two-step sample stratified at marz level was developed. All marzes, as well as all urban and rural settlements were included in the sample population according to the share of population residing in those settlements as percent to the total population in the country. In the first step, the settlements, i.e. primary sample units, were selected: 43 towns out of 48 or 90 percent of all towns in Armenia were surveyed during the year; also 216 villages out of 951 or 23 percent of all villages in the country were covered by the survey. In the second step, the respondent households were selected: 6,816 households (5,088 from urban and 1,728 from rural settlements). As a result, for the first time since 1996 survey data were representative at the marz level.

During the 2007-2012 surveys the sampling frame for ILCS was designed according to the database of addresses for the 2001 Population Census, which was developed with the World Bank technical assistance. The sample consisted of two parts: core sample and oversample.

1) For the creation of core sample, the sample frame (database of addresses of all households in Armenia) was divided into 48 strata including 12 communities of Yerevan city. The households from other regions (marzes) were grouped according to three categories: large towns (with population of 15000 and higher), villages and other towns. Large towns formed by 16 groups (strata), while the villages and towns formed by 10 strata each. According to that division, a random, two-step sample stratified at the marz level was developed. All marzes, as well as all urban and rural settlements were included in the sample population according to the share of households residing in those settlements as percent to the total households in the country. In the first step, using the PPS method the enumeration units (i.e., primary sample units to be surveyed during the year) were selected. 2007 sample includes 48 urban and 18 rural enumeration areas per month. 2) The oversample was drawn from the list of villages included in MCA-Armenia Rural Roads Rehabilitation Project. The enumeration areas of villages that were already in the core sample were excluded from that list. From the remaining enumeration areas 18 enumeration areas were selected per month. Thus, the rural sample size was doubled. 3) After merging the core sample and oversample, the survey households were selected in the second step. 656 households were surveyed per month, from which 368 from urban and 288 from rural settlements. Each month 82 interviewers had conducted field work, and their workload included 8 households per month. In 2007 number of surveyed households was 7,872 (4,416 from urban and 3,456 from rural areas).

For the survey 2013 the sample frame for ILCS was designed in accordance with the database of addresses of all private households in the country developed on basis of the 2001 Population Census results, with the technical assistance of the World Bank. The method of systematic representative probability sampling was used to frame the sample. For the purpose of drawing the sample, the sample frame was divided into 32 strata including 12 communities of Yerevan City (currently, the administrative districts). According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all urban and rural communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration areas - that is primary sample units to be surveyed during the year - were selected. The ILCS 2013 sample included 32 enumeration areas in urban and 16 enumeration areas in rural communities per month. The households to be surveyed were selected in the second round. A total of 432 households were surveyed per month, of which 279 and 153 households from urban and rural communities, respectively. Every month 48 interviewers went on field work with a workload of 9 households per month.

The sample frame for 2014-2016 was designed in accordance with the database of addresses of all private households in the country developed on basis of the 2011 Population Census results, with the technical assistance of the World Bank. The method of systematic representative probability sampling was used to frame the sample.
For drawing the sample, the sample frame was divided into 32 strata including 12 communities of Yerevan City (currently, the administrative districts). According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all urban and rural communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration areas - that is primary sample units to be surveyed during the year - were selected. The ILCS 2014 sample included 30 enumeration areas in urban and 18 enumeration areas in rural communities per month. The method of representative probability sampling was used to frame the sample. At regional level, all communities were grouped into two categories - towns and villages. According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all rural and urban communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration districts - that is primary sample units to be surveyed during the year - were selected. The ILCS 2015 sample included 30 enumeration districts in urban and 18 enumeration districts in rural communities per month.

Mode of data collection

Face-to-face [f2f]

Research instrument

The Questionnaire is filled in by the interviewer during the least five visits to households per month. During face-to-face interviews with the household head or another knowledgeable adult member, the interviewer collects information on the composition and housing conditions of the household, the employment status, educational level and health condition of the members, availability and use of land, livestock, and agricultural machinery, monetary and commodity flows between households, and other information.

The 2015 survey questionnaire had the following sections: (1) "List of Household Members", (2) "Migration", (3) "Housing and Dwelling Conditions", (4) "Employment", (5) "Education", (6) "Agriculture", (7) "Food Production", (8) "Monetary and Commodity Flows between Households", (9) "Health (General) and Healthcare", (10) "Debts", (11) "Subjective Assessment of Living Conditions", (12) "Provision of Services", (13) "Social Assistance", (14) "Households as Employers for Service Personnel", and (15) "Household Monthly Consumption of Energy Resources".

The Diary is completed directly by the household for one month. Every day the household would record all its expenditures on food, non-food products and services, also giving a detailed description of such purchases; e.g. for food products the name, quantity, cost, and place of purchase of the product is recorded. Besides, the household records its consumption of food products received and used from its own land and livestock, as well as from other sources (e.g. gifts, humanitarian aid). Non-food products and services purchased or received for free are also recorded in the diary. Then, the household records its income received during the month. At the end of the month, information on rarely used food products, durable goods and ceremonies is recorded, as well. The records in the diary are verified by the interviewer in the course of 5

Chinese household panel data 2011-2015 covering all CPG purchases, including support files barcode, shopcode, panelist (including information on age, household size, income, city (tier)). In the current study, we want to shed light on the role of brand price as a marketing mix instrument for consumer packaged goods (CPG) in a dynamic emerging market like China. Specifically, we intend to answer the following research questions: What is the relationship between a brand’s price and market share over time: How price sensitive are Chinese consumers for CPG products, and how do CPG sellers change price in response to a change in market share? Do these relations differ for the short versus long term? What brand and category contexts moderate these relations? To empirically address these issues, we aim to conduct a comprehensive study on the link between prices and market shares across hundreds of brands in a large set of packaged goods over a five-year period.

Abstract

The Integrated Survey of Living Standards (ISLS), renamed in 2004 to Integrated Survey of Living Conditions Survey (ILCS) is conducted annually by the NSS National Statistical Service of the Republic of Armenia, formed the basis for monitoring living conditions in Armenia. The ILCS is a universally recognized best-practice survey for collecting data to inform about the living standards of households. The ILCS comprises comprehensive and valuable data on the welfare of households and separate individuals which gives the NSS an opportunity to provide the public with up to date information on the population’s income, expenditures, the level of poverty and the other changes in living standards on an annual basis.

Geographic coverage

Urban and rural communities

Analysis unit

• Households;
• Individuals.

Kind of data

Sample survey data [ssd]

Sampling procedure

During the 2001-2003 surveys two-stage random sample was used; the first stage covered the selection of settlements - cities and villages, while the second stage was focused on the selection of households in these settlements. The surveys were conducted on the principle of monthly rotation of households by clusters (sample units). In 2002 and 2003 the number of households was 387 with the sample covering 14 cities and 30 villages in 2002 and 17 cities and 20 villages in 2003.

During the 2004-2006 surveys the sampling frame for the ILCS was built using the database of addresses for the 2001 Population Census; the database was developed with the World Bank technical assistance. The database of addresses of all households in Armenia was divided into 48 strata including 12 communities of Yerevan city. The households from other regions (marzes) were grouped according to the following three categories: big towns with 15,000 and more population; villages, and other towns. Big towns formed 16 strata (the only exception was the Vayots Dzor marz where there are no big towns). The villages and other towns formed 10 strata each. According to this division, a random, two-step sample stratified at marz level was developed. All marzes, as well as all urban and rural settlements were included in the sample population according to the share of population residing in those settlements as percent to the total population in the country. In the first step, the settlements, i.e. primary sample units, were selected: 43 towns out of 48 or 90 percent of all towns in Armenia were surveyed during the year; also 216 villages out of 951 or 23 percent of all villages in the country were covered by the survey. In the second step, the respondent households were selected: 6,816 households (5,088 from urban and 1,728 from rural settlements). As a result, for the first time since 1996 survey data were representative at the marz level.

During the 2007-2012 surveys the sampling frame for ILCS was designed according to the database of addresses for the 2001 Population Census, which was developed with the World Bank technical assistance. The sample consisted of two parts: core sample and oversample.

1) For the creation of core sample, the sample frame (database of addresses of all households in Armenia) was divided into 48 strata including 12 communities of Yerevan city. The households from other regions (marzes) were grouped according to three categories: large towns (with population of 15000 and higher), villages and other towns. Large towns formed by 16 groups (strata), while the villages and towns formed by 10 strata each. According to that division, a random, two-step sample stratified at the marz level was developed. All marzes, as well as all urban and rural settlements were included in the sample population according to the share of households residing in those settlements as percent to the total households in the country. In the first step, using the PPS method the enumeration units (i.e., primary sample units to be surveyed during the year) were selected. 2007 sample includes 48 urban and 18 rural enumeration areas per month. 2) The oversample was drawn from the list of villages included in MCA-Armenia Rural Roads Rehabilitation Project. The enumeration areas of villages that were already in the core sample were excluded from that list. From the remaining enumeration areas 18 enumeration areas were selected per month. Thus, the rural sample size was doubled. 3) After merging the core sample and oversample, the survey households were selected in the second step. 656 households were surveyed per month, from which 368 from urban and 288 from rural settlements. Each month 82 interviewers had conducted field work, and their workload included 8 households per month. In 2007 number of surveyed households was 7,872 (4,416 from urban and 3,456 from rural areas).

For the survey 2013 the sample frame for ILCS was designed in accordance with the database of addresses of all private households in the country developed on basis of the 2001 Population Census results, with the technical assistance of the World Bank. The method of systematic representative probability sampling was used to frame the sample. For the purpose of drawing the sample, the sample frame was divided into 32 strata including 12 communities of Yerevan City (currently, the administrative districts). According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all urban and rural communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration areas - that is primary sample units to be surveyed during the year - were selected. The ILCS 2013 sample included 32 enumeration areas in urban and 16 enumeration areas in rural communities per month. The households to be surveyed were selected in the second round. A total of 432 households were surveyed per month, of which 279 and 153 households from urban and rural communities, respectively. Every month 48 interviewers went on field work with a workload of 9 households per month.

The sample frame for 2014-2016 was designed in accordance with the database of addresses of all private households in the country developed on basis of the 2011 Population Census results, with the technical assistance of the World Bank. The method of systematic representative probability sampling was used to frame the sample.
For drawing the sample, the sample frame was divided into 32 strata including 12 communities of Yerevan City (currently, the administrative districts). According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all urban and rural communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration areas - that is primary sample units to be surveyed during the year - were selected. The ILCS 2014 sample included 30 enumeration areas in urban and 18 enumeration areas in rural communities per month. The method of representative probability sampling was used to frame the sample. At regional level, all communities were grouped into two categories - towns and villages. According to this division, a two-tier sample was drawn stratified by regions and by Yerevan. All regions and Yerevan, as well as all rural and urban communities were included in the sample in accordance to the shares of their resident households within the total number of households in the country. In the first round, enumeration districts - that is primary sample units to be surveyed during the year - were selected. The ILCS 2015 sample included 30 enumeration districts in urban and 18 enumeration districts in rural communities per month.

Mode of data collection

Face-to-face [f2f]

Research instrument

The Questionnaire is filled in by the interviewer during the least five visits to households per month. During face-to-face interviews with the household head or another knowledgeable adult member, the interviewer collects information on the composition and housing conditions of the household, the employment status, educational level and health condition of the members, availability and use of land, livestock, and agricultural machinery, monetary and commodity flows between households, and other information.

The 2003 survey questionnaire had the following sections: (1) "List of Household Members", (2) "Housing Facilities", (3) "Migration", (4) "Education", (5) "Agriculture", (6) "Monetary and Commodity Flows between Households", (7) "Health (General) and Healthcare", (8) "Savings and Debts", (9) "Social Assistance"

The Diary is completed directly by the household for one month. Every day the household would record all its expenditures on food, non-food products and services, also giving a detailed description of such purchases; e.g. for food products the name, quantity, cost, and place of purchase of the product is recorded. Besides, the household records its consumption of food products received and used from its own land and livestock, as well as from other sources (e.g. gifts, humanitarian aid). Non-food products and services purchased or received for free are also recorded in the diary. Then, the household records its income received during the month. At the end of the month, information on rarely used food products, durable goods and ceremonies is recorded, as well. The records in the diary are verified by the interviewer in the course of 5 mandatory visits to the household during the survey month.

The Survey Diary has the following sections: (1) food purchased during the day, (2) food consumed at home

Chinese household panel data 2011-2016 covering all CPG purchases, including support files barcode, shopcode, panelist (incl info on age, household size, income, city (tier)). Research questions: How do brand sales in the Chinese market change as the fraction of groceries sold online goes up? What category- and brand related factors influence this change, and what is the direction of the effect? How can managers use these insights to preserve and improve their brand performance in an increasingly electronic grocery market?

For detailed information, visit the Tucson Equity Priority Index StoryMap.Download the Data DictionaryWhat is the Tucson Equity Priority Index (TEPI)?The Tucson Equity Priority Index (TEPI) is a tool that describes the distribution of socially vulnerable demographics. It categorizes the dataset into 5 classes that represent the differing prioritization needs based on the presence of social vulnerability: Low (0-20), Low-Moderate (20-40), Moderate (40-60), Moderate-High (60-80) High (80-100). Each class represents 20% of the dataset’s features in order of their values. The features within the Low (0-20) classification represent the areas that, when compared to all other locations in the study area, have the lowest need for prioritization, as they tend to have less socially vulnerable demographics. The features that fall into the High (80-100) classification represent the 20% of locations in the dataset that have the greatest need for prioritization, as they tend to have the highest proportions of socially vulnerable demographics. How is social vulnerability measured?The Tucson Equity Priority Index (TEPI) examines the proportion of vulnerability per feature using 11 demographic indicators:Income Below Poverty: Households with income at or below the federal poverty level (FPL), which in 2023 was $14,500 for an individual and $30,000 for a family of fourUnemployment: Measured as the percentage of unemployed persons in the civilian labor forceHousing Cost Burdened: Homeowners who spend more than 30% of their income on housing expenses, including mortgage, maintenance, and taxesRenter Cost Burdened: Renters who spend more than 30% of their income on rentNo Health Insurance: Those without private health insurance, Medicare, Medicaid, or any other plan or programNo Vehicle Access: Households without automobile, van, or truck accessHigh School Education or Less: Those highest level of educational attainment is a High School diploma, equivalency, or lessLimited English Ability: Those whose ability to speak English is "Less Than Well."People of Color: Those who identify as anything other than Non-Hispanic White Disability: Households with one or more physical or cognitive disabilities Age: Groups that tend to have higher levels of vulnerability, including children (those below 18), and seniors (those 65 and older)An overall percentile value is calculated for each feature based on the total proportion of the above indicators in each area. How are the variables combined?These indicators are divided into two main categories that we call Thematic Indices: Economic and Personal Characteristics. The two thematic indices are further divided into five sub-indices called Tier-2 Sub-Indices. Each Tier-2 Sub-Index contains 2-3 indicators. Indicators are the datasets used to measure vulnerability within each sub-index. The variables for each feature are re-scaled using the percentile normalization method, which converts them to the same scale using values between 0 to 100. The variables are then combined first into each of the five Tier-2 Sub-Indices, then the Thematic Indices, then the overall TEPI using the mean aggregation method and equal weighting. The resulting dataset is then divided into the five classes, where:High Vulnerability (80-100%): Representing the top classification, this category includes the highest 20% of regions that are the most socially vulnerable. These areas require the most focused attention. Moderate-High Vulnerability (60-80%): This upper-middle classification includes areas with higher levels of vulnerability compared to the median. While not the highest, these areas are more vulnerable than a majority of the dataset and should be considered for targeted interventions. Moderate Vulnerability (40-60%): Representing the middle or median quintile, this category includes areas of average vulnerability. These areas may show a balanced mix of high and low vulnerability. Detailed examination of specific indicators is recommended to understand the nuanced needs of these areas. Low-Moderate Vulnerability (20-40%): Falling into the lower-middle classification, this range includes areas that are less vulnerable than most but may still exhibit certain vulnerable characteristics. These areas typically have a mix of lower and higher indicators, with the lower values predominating. Low Vulnerability (0-20%): This category represents the bottom classification, encompassing the lowest 20% of data points. Areas in this range are the least vulnerable, making them the most resilient compared to all other features in the dataset.