Mindsets and Segmentation: Promoting 2020 Census self-response Laura Kail, PSB Gina Walejko, U.S. Census Bureau Brian Kriz, PSB Robert Kulzick, PSB Shawnna Mullenax, PSB Hubert Shang, PSB July 12, 2019 The views and opinions expressed in this presentation are those of the authors and 2020 CBAMS PUMS DRB#: CBDRB-FY18-422 do not necessarily reflect the official policy or position of the U.S. Census Bureau. 2020 ICC Modeled Scores DRB#: CBDRB-FY18-311
Introduction • Segmentation is frequently used to differentiate potential buyers of products or services • Profiles of customers are often created to help with ad campaigns and creative message development • In this case, every person in the United States is the customer, and the product is the 2020 Census • Mindsets and segmentation will help communication experts tailor advertisements and messaging strategies to have the greatest possible impact 2
Goals M INDSETS M INDSETS T RACT S EGMENTS T RACT S EGMENTS • • Cluster individuals into cohesive Cluster tracts into cohesive groups (i.e. mindsets) with groups (i.e. segments) with similar attitudes toward and similar self-response and knowledge about the 2020 demographics Census • Have tract segments that are • Have mindset groupings that distinct from each other are distinct from each other 3
Input Data Sources M INDSETS T RACT S EGMENTS • • Responses to 50+ non- Modeled scores predicting demographic questions from overall self-response rates and the 2020 CBAMS the proportion of self-response that will be online • 33 tract-level demographic characteristics from Census Bureau sources (2010 Census and 2016 ACS) 4
Mindsets and Segmentation Process 1. Select inputs 2. Reduce inputs to a smaller number of factors using Principal Component Analysis (PCA) 3. Group individuals/tracts into 14 candidate solutions 4. Narrow candidates to 3 solutions each 5. Communication experts select a final mindset and final segment solution 6. Name the mindsets/segments and develop descriptive profiles of each 7. Use the mindsets/segments to inform the communication campaign 5
Selection Criteria M INDSETS T RACT S EGMENTS • • Differentiation on stated intent to Differentiation on the predicted self- participate in the 2020 Census response rates • • Balanced distribution of mindset Balanced distribution of segment sizes sizes • • Meaningful demographic Meaningful demographic characteristics and attitudes toward characteristics, mindsets, and media and knowledge about the census consumption habits 6
Mindset Descriptions Eager Engagers Fence Sitters Confidentiality Minded 19% of U.S. Population 32% of U.S. Population 15% of U.S. Population 82% Intent to Respond 71% Intent to Respond 63% Intent to Respond • Highest intent to respond. • Above-average intent to respond. • Most concerned answers will be used against them. • Highest levels of civic engagement. • Highest percentage White. • Highest percentage foreign-born. • Highest knowledge about the census. • Highest percentage male. • Slightly below average intent to respond. Head Nodders Wary Skeptics Disconnected Doubters 9% of U.S. Population 14% of U.S. Population 10% of U.S. Population 60% Intent to Respond Intent to Respond Intent to Respond 59% 51% • Most likely to respond “Yes” to all • Lowest trust in government. • Lowest intent to respond. knowledge questions. • Highest apathy about the census. • Lowest frequency of internet use. • Above-average percentage of • High percentage of Black/African- • Highest percentage of people ages foreign-born people. Americans. 65 years and older. • Highest percentage of people ages 18 to 34. 7 Note: U.S. population percentages do not add to 100% due to rounding error.
Mindset Profiles 8
Tract Segment Descriptions Responsive Suburbia Main Street Middle Country Roads Downtown Dynamic 67% 60% 71% 24% 21% 16% 59% 9% Of the U.S. Of the U.S. Predicted Of the U.S. Predicted Of the U.S. Predicted Predicted Population Population Population Population Self-Response Self-Response Self-Response Self-Response • • • • High predicted rate of response, with a high High predicted rate of response, with an average Slightly below-average predicted rate of response, Slightly below-average predicted rate of response, percentage of that response coming online. percentage of that response coming online. with a below-average percentage of that response with a high percentage of that response coming coming online. online. • • Found in suburban neighborhoods of single-family Found in small towns and less densely populated • • homes. areas surrounding urban centers. Found in rural areas predominantly in the eastern Found in densely populated metro centers. United States, surrounding small towns and • • • High % college educated, high % married, and high Low diversity and a higher % 65 or older than the High % college educated, above-average % foreign- outside the suburbs of major cities. median household incomes. national average. born, high % 25-44 compared to the nation as a • High % owner-occupied housing, low % college whole, and high median household incomes. educated, and below-average median household incomes. Student and Military Rural Delta and Multicultural Mosaic Sparse Spaces Communities Urban Enclaves 56% 2% 45% 7% 49% 5% 43% 14% Of the U.S. Of the U.S. Of the U.S. Of the U.S. Predicted Predicted Predicted Predicted Population Self-Response Population Self-Response Population Self-Response Population Self-Response • • • • Below-average predicted rate of response, with a Below-average predicted rate of response, with Low predicted rate of response, with a below- Low predicted rate of response, with the lowest high percentage of that response coming online. below-average internet response. average percentage of that response coming percentage of that response coming online. online. • • • Found in communities around college campuses or Found in rural areas predominantly in the western Found in rural parts of the southeastern United • military bases. United States, Appalachia, northern Maine, and Found in California’s Central Valley and parts of States, as well as concentrations in urban areas. Michigan’s Upper Peninsula. New Mexico, Texas, Florida, as well as • • Low % college educated, low median household A majority 18-24, high % college educated, and concentrations in urban areas. • high % renter-occupied housing. High % owner-occupied housing and below- incomes, below-average levels of internet access, • average levels of internet access. High % foreign-born, low % college educated, and and majority non-Hispanic African American. majority Hispanic. 9 Note: U.S. population percentages do not add up to 100% due to tracts with no ACS mailout and, therefore, no tract segment assigned.
National View of Tract Segmentation Responsive Suburbia Downtown Dynamic Multicultural Mosaic Main Street Middle Student and Military Communities Rural Delta and Urban Enclaves Country Roads Sparse Spaces No ACS Mailout 10
Washington, D.C. View of Tract Segmentation Responsive Suburbia Main Street Middle Country Roads Downtown Dynamic Student and Military Communities Sparse Spaces Multicultural Mosaic Rural Delta and Urban Enclaves No ACS Mailout 11
Mindset Composition of Tract Segments Responsive Suburbia Main Street Middle Country Roads Downtown Dynamic Eager Eager Eager 18% 16% 23% Eager 21% Engagers Engagers Engagers Engagers Fence 33% Fence Fence 33% 29% Fence 40% Sitters Sitters Sitters Sitters Confidentiality Confidentiality Confidentiality 14% 14% 17% Confidentiality 15% Minded Minded Minded Minded Head Head Head 10% 7% 12% Head 9% Nodders Nodders Nodders Nodders Wary 14% Wary 15% Wary 13% Wary 10% Skeptics Skeptics Skeptics Skeptics Disconnected 11% Disconnected Disconnected 15% 6% Disconnected 5% Doubters Doubters Doubters Doubters Rural Delta and Student and Military Sparse Spaces Multicultural Mosaic Urban Enclaves Communities Eager Eager 32% Eager 11% Eager 16% 18% Engagers Engagers Engagers Engagers Fence Fence Fence 37% Fence 16% 13% 37% Sitters Sitters Sitters Sitters Confidentiality Confidentiality Confidentiality Confidentiality 33% 22% 8% 12% Minded Minded Minded Minded Head Head Head Head 11% 11% 11% 6% Nodders Nodders Nodders Nodders Wary Wary Wary 20% Wary 22% 11% 15% Skeptics Skeptics Skeptics Skeptics Disconnected Disconnected Disconnected 2% Disconnected 9% 17% 11% Doubters Doubters Doubters Doubters National Percentage Segment Percentage Note: Mindset percentages within some segments do not add up to 100% due to rounding. 12
Tract Segment Profiles 13
Applications Mindsets and segmentation will be used to inform the communications campaign in the following ways: Overall Strategy Messaging Partnerships 14
Public Use www.census.gov/roam 15
Questions? Laura Kail, PSB lkail@ps-b.com Gina Walejko, U.S. Census Bureau Brian Kriz, PSB Robert Kulzick, PSB Shawnna Mullenax, PSB Hubert Shang, PSB
A PPENDIX
Recommend
More recommend
