In 2023, just over 50 percent of Americans had an annual household income that was less than 75,000 U.S. dollars. The median household income was 80,610 U.S. dollars in 2023. Income and wealth in the United States After the economic recession in 2009, income inequality in the U.S. is more prominent across many metropolitan areas. The Northeast region is regarded as one of the wealthiest in the country. Maryland, New Jersey, and Massachusetts were among the states with the highest median household income in 2020. In terms of income by race and ethnicity, the average income of Asian households was 94,903 U.S. dollars in 2020, while the median income for Black households was around half of that figure. What is the U.S. poverty threshold? The U.S. Census Bureau annually updates its list of poverty levels. Preliminary estimates show that the average poverty threshold for a family of four people was 26,500 U.S. dollars in 2021, which is around 100 U.S. dollars less than the previous year. There were an estimated 37.9 million people in poverty across the United States in 2021, which was around 11.6 percent of the population. Approximately 19.5 percent of those in poverty were Black, while 8.2 percent were white.

Annual descriptive price statistics for each calendar year 2005 – 2023 for 11 Local Government Districts in Northern Ireland. The statistics include: • Minimum sale price • Lower quartile sale price • Median sale price • Simple Mean sale price • Upper Quartile sale price • Maximum sale price • Number of verified sales Prices are available where at least 30 sales were recorded in the area within the calendar year which could be included in the regression model i.e. the following sales are excluded: • Non Arms-Length sales • sales of properties where the habitable space are less than 30m2 or greater than 1000m2 • sales less than £20,000. Annual median or simple mean prices should not be used to calculate the property price change over time. The quality (where quality refers to the combination of all characteristics of a residential property, both physical and locational) of the properties that are sold may differ from one time period to another. For example, sales in one quarter could be disproportionately skewed towards low-quality properties, therefore producing a biased estimate of average price. The median and simple mean prices are not ‘standardised’ and so the varying mix of properties sold in each quarter could give a false impression of the actual change in prices. In order to calculate the pure property price change over time it is necessary to compare like with like, and this can only be achieved if the ‘characteristics-mix’ of properties traded is standardised. To calculate pure property change over time please use the standardised prices in the NI House Price Index Detailed Statistics file.

Annual descriptive price statistics for each calendar year 2005 – 2023 for 462 electoral wards within 11 Local Government Districts.

The statistics include:

• Minimum sale price

• Lower quartile sale price

• Median sale price

• Simple Mean sale price

• Upper Quartile sale price

• Maximum sale price

• Number of verified sales

Prices are available where at least 30 sales were recorded in the area within the calendar year which could be included in the regression model i.e. the following sales are excluded:

• Non Arms-Length sales

• sales of properties where the habitable space are less than 30m2 or greater than 1000m2

• sales less than £20,000.

Annual median or simple mean prices should not be used to calculate the property price change over time.
The quality (where quality refers to the combination of all characteristics of a residential property, both physical and locational) of the properties that are sold may differ from one time period to another. For example, sales in one quarter could be disproportionately skewed towards low-quality properties, therefore producing a biased estimate of average price. The median and simple mean prices are not ‘standardised’ and so the varying mix of properties sold in each quarter could give a false impression of the actual change in prices. In order to calculate the pure property price change over time it is necessary to compare like with like, and this can only be achieved if the ‘characteristics-mix’ of properties traded is standardised. To calculate pure property change over time please use the standardised prices in the NI House Price Index Detailed Statistics file.

U.S. citizens with a professional degree had the highest median household income in 2023, at 172,100 U.S. dollars. In comparison, those with less than a 9th grade education made significantly less money, at 35,690 U.S. dollars. Household income The median household income in the United States has fluctuated since 1990, but rose to around 70,000 U.S. dollars in 2021. Maryland had the highest median household income in the United States in 2021. Maryland’s high levels of wealth is due to several reasons, and includes the state's proximity to the nation's capital. Household income and ethnicity The median income of white non-Hispanic households in the United States had been on the rise since 1990, but declining since 2019. While income has also been on the rise, the median income of Hispanic households was much lower than those of white, non-Hispanic private households. However, the median income of Black households is even lower than Hispanic households. Income inequality is a problem without an easy solution in the United States, especially since ethnicity is a contributing factor. Systemic racism contributes to the non-White population suffering from income inequality, which causes the opportunity for growth to stagnate.

State politics researchers commonly employ ordinary least squares (OLS) regression or one of its variants to test linear hypotheses. However, OLS is easily influenced by outliers and thus can produce misleading results when the error term distribution has heavy tails. Here we demonstrate that median regression (MR), an alternative to OLS that conditions the median of the dependent variable (rather than the mean) on the independent variables, can be a solution to this problem. Then we propose and validate a hypothesis test that applied researchers can use to select between OLS and MR in a given sample of data. Finally, we present two examples from state politics research in which (1) the test selects MR over OLS and (2) differences in results between the two methods could lead to different substantive inferences. We conclude that MR and the test we propose can improve linear models in state politics research.

This is the unadjusted median house priced for residential property sales (transactions) in the area for a 12 month period with April in the middle (year-ending September). These figures have been produced by the ONS (Office for National Statistics) using the Land Registry (LR) Price Paid data on residential dwelling transactions.

The LR Price Paid data are comprehensive in that they capture changes of ownership for individual residential properties which have sold for full market value and covers both cash sales and those involving a mortgage.

The median is the value determined by putting all the house sales for a given year, area and type in order of price and then selecting the price of the house sale which falls in the middle. The median is less susceptible to distortion by the presence of extreme values than is the mean. It is the most appropriate average to use because it best takes account of the skewed distribution of house prices.

Note that a transaction occurs when a change of freeholder or leaseholder takes place regardless of the amount of money involved and a property can transact more than once in the time period.

The LR records the actual price for which the property changed hands. This will usually be an accurate reflection of the market value for the individual property, but it is not always the case. In order to generate statistics that more accurately reflect market values, the LR has excluded records of houses that were not sold at market value from the dataset. The remaining data are considered a good reflection of market values at the time of the transaction. For full details of exclusions and more information on the methodology used to produce these statistics please see http://www.ons.gov.uk/peoplepopulationandcommunity/housing/qmis/housepricestatisticsforsmallareasqmi

The LR Price Paid data are not adjusted to reflect the mix of houses in a given area. Fluctuations in the types of house that are sold in that area can cause differences between the median transactional value of houses and the overall market value of houses. Therefore these statistics differ to the new UK House Price Index (HPI) which reports mix-adjusted average house prices and house price indices.

If, for a given year, for house type and area there were fewer than 5 sales records in the LR Price Paid data, the house price statistics are not reported. Data is Powered by LG Inform Plus and automatically checked for new data on the 3rd of each month.

VITAL SIGNS INDICATOR Jobs by Wage Level (EQ1)

FULL MEASURE NAME Distribution of jobs by low-, middle-, and high-wage occupations

LAST UPDATED January 2019

DESCRIPTION Jobs by wage level refers to the distribution of jobs by low-, middle- and high-wage occupations. In the San Francisco Bay Area, low-wage occupations have a median hourly wage of less than 80% of the regional median wage; median wages for middle-wage occupations range from 80% to 120% of the regional median wage, and high-wage occupations have a median hourly wage above 120% of the regional median wage.

DATA SOURCE California Employment Development Department OES (2001-2017) http://www.labormarketinfo.edd.ca.gov/data/oes-employment-and-wages.html

American Community Survey (2001-2017) http://api.census.gov

CONTACT INFORMATION vitalsigns.info@bayareametro.gov

METHODOLOGY NOTES (across all datasets for this indicator) Jobs are determined to be low-, middle-, or high-wage based on the median hourly wage of their occupational classification in the most recent year. Low-wage jobs are those that pay below 80% of the regional median wage. Middle-wage jobs are those that pay between 80% and 120% of the regional median wage. High-wage jobs are those that pay above 120% of the regional median wage. Regional median hourly wages are estimated from the American Community Survey and are published on the Vital Signs Income indicator page. For the national context analysis, occupation wage classifications are unique to each metro area. A low-wage job in New York, for instance, may be a middle-wage job in Miami. For the Bay Area in 2017, the median hourly wage for low-wage occupations was less than $20.86 per hour. For middle-wage jobs, the median ranged from $20.86 to $31.30 per hour; and for high-wage jobs, the median wage was above $31.30 per hour.

Occupational employment and wage information comes from the Occupational Employment Statistics (OES) program. Regional and subregional data is published by the California Employment Development Department. Metro data is published by the Bureau of Labor Statistics. The OES program collects data on wage and salary workers in nonfarm establishments to produce employment and wage estimates for some 800 occupations. Data from non-incorporated self-employed persons are not collected, and are not included in these estimates. Wage estimates represent a three-year rolling average.

Due to changes in reporting during the analysis period, subregion data from the EDD OES have been aggregated to produce geographies that can be compared over time. West Bay is San Mateo, San Francisco, and Marin counties. North Bay is Sonoma, Solano and Napa counties. East Bay is Alameda and Contra Costa counties. South Bay is Santa Clara County from 2001-2004 and Santa Clara and San Benito counties from 2005-2017.

Due to changes in occupation classifications during the analysis period, all occupations have been reassigned to 2010 SOC codes. For pre-2009 reporting years, all employment in occupations that were split into two or more 2010 SOC occupations are assigned to the first 2010 SOC occupation listed in the crosswalk table provided by the Census Bureau. This method assumes these occupations always fall in the same wage category, and sensitivity analysis of this reassignment method shows this is true in most cases.

In order to use OES data for time series analysis, several steps were taken to handle missing wage or employment data. For some occupations, such as airline pilots and flight attendants, no wage information was provided and these were removed from the analysis. Other occupations did not record a median hourly wage (mostly due to irregular work hours) but did record an annual average wage. Nearly all these occupations were in education (i.e. teachers). In this case, a 2080 hour-work year was assumed and [annual average wage/2080] was used as a proxy for median income. Most of these occupations were classified as high-wage, thus dispelling concern of underestimating a median wage for a teaching occupation that requires less than 2080 hours of work a year (equivalent to 12 months fulltime). Finally, the OES has missing employment data for occupations across the time series. To make the employment data comparable between years, gaps in employment data for occupations are ‘filled-in’ using linear interpolation if there are at least two years of employment data found in OES. Occupations with less than two years of employment data were dropped from the analysis. Over 80% of interpolated cells represent missing employment data for just one year in the time series. While this interpolating technique may impact year-over-year comparisons, the long-term trends represented in the analysis generally are accurate.

Graph and download economic data for Real Median Personal Income in the United States (MEPAINUSA672N) from 1974 to 2023 about personal income, personal, income, median, real, and USA.

1. The performance of ectotherms integrated over time depends in part on the position and shape of the distribution of body temperatures (Tb) experienced during activity. For several complementary reasons, physiological ecologists have long expected that Tb distributions during activity should have a long left tail (left-skewed); but only infrequently have they quantified the magnitude and direction of Tb skewness in nature.
2. To evaluate whether left-skewed Tb distributions are general for diurnal desert lizards, we compiled and analyzed Tb (∑ = 9,023 temperatures) from our own prior studies of active desert lizards on three continents (25 species in Western Australia, 10 in the Kalahari Desert of Africa, and 10 species in western North America). We gathered these data over several decades, using standardized techniques.
3. Many species showed significantly left-skewed Tb distributions, even when records were restricted to summer months. However, magnitudes of skewness were always small, such that mean Tb were never more than 1°C lower than median Tb. The significance of Tb skewness was sensitive to sample size, and power tests reinforced this sensitivity.
4. The magnitude of skewness was not obviously related to phylogeny, desert, body size, or median body temperature. Moreover, formal phylogenetic analysis is inappropriate because geography and phylogeny are confounded (that is, are highly collinear).
5. Skewness might be limited if lizards pre-warm inside retreats before emerging in the morning, emerge only when operative temperatures are high enough to speed warming to activity Tb, or if cold lizards are especially wary and difficult to spot or catch. Telemetry studies may help evaluate these possibilities.

Data Simulation We simulate performance data for both women and men across three aspects of multitasking:

Task Switching Speed: The speed at which individuals can switch between tasks. Attention Allocation: The efficiency in distributing attention across multiple tasks. Task Completion Time: The time taken to complete tasks. For each aspect, performance data for women and men are generated using normal distributions with specified means and standard deviations:

Task Switching Speed: Women (mean = 0.8, std dev = 0.1), Men (mean = 0.6, std dev = 0.1) Attention Allocation: Women (mean = 0.8, std dev = 0.1), Men (mean = 0.6, std dev = 0.1) Task Completion Time: Women (mean = 10, std dev = 2), Men (mean = 12, std dev = 2) Summary Statistics Summary statistics for each aspect are calculated to provide insights into the central tendencies and variability of performance for both genders. These include the mean, median, and standard deviation.

Task Switching Speed:

Women: Mean = 0.808, Median = 0.822, Std Dev = 0.097 Men: Mean = 0.600, Median = 0.593, Std Dev = 0.100 Attention Allocation:

Women: Mean = 0.823, Median = 0.814, Std Dev = 0.101 Men: Mean = 0.607, Median = 0.613, Std Dev = 0.103 Task Completion Time:

Women: Mean = 10.135, Median = 10.153, Std Dev = 2.129 Men: Mean = 11.687, Median = 11.500, Std Dev = 1.806 These statistics are used to compare the performance of women and men, highlighting differences in their multitasking abilities.

Hypothesis Testing To determine if the observed differences in performance are statistically significant, t-tests are conducted for each aspect. The p-values obtained from these tests indicate the significance of the differences:

Task Switching Speed: p-value = 1.294e-17 Attention Allocation: p-value = 1.249e-17 Task Completion Time: p-value = 0.000183 A p-value less than 0.05 suggests that the differences are statistically significant, meaning that the observed performance differences between women and men are unlikely due to random chance.

Graphical Representations Histograms are created for each aspect to visualize the distribution of performance scores for women and men. These graphs provide a clear and intuitive understanding of the differences in multitasking performance.

The median waiting time is the middle value when all waits are ordered from shortest to longest, meaning half of all current waits are less than the median and half are more than the median. It is commonly used in preference to the mean as it is less influenced by extreme values.

Dataset contains counts and measures for families and extended families from the 2013, 2018, and 2023 Censuses. Data is available by statistical area 2.

​

The variables included in this dataset are for families and extended families in households in occupied private dwellings:

• Count of families
• Family type
• Number of people in family
• Average number of people in family
• Total family income
• Median ($) total family income
• Count of extended families
• Extended family type
• Total extended family income
• Median ($) total extended family income.

​

Download lookup file from Stats NZ ArcGIS Online or embedded attachment in Stats NZ geographic data service. Download data table (excluding the geometry column for CSV files) using the instructions in the Koordinates help guide.

​

Footnotes

Geographical boundaries

Statistical standard for geographic areas 2023 (updated December 2023) has information about geographic boundaries as of 1 January 2023. Address data from 2013 and 2018 Censuses was updated to be consistent with the 2023 areas. Due to the changes in area boundaries and coding methodologies, 2013 and 2018 counts published in 2023 may be slightly different to those published in 2013 or 2018.

​

Caution using time series

Time series data should be interpreted with care due to changes in census methodology and differences in response rates between censuses. The 2023 and 2018 Censuses used a combined census methodology (using census responses and administrative data), while the 2013 Census used a full-field enumeration methodology (with no use of administrative data).

​

About the 2023 Census dataset

For information on the 2023 dataset see Using a combined census model for the 2023 Census. We combined data from the census forms with administrative data to create the 2023 Census dataset, which meets Stats NZ's quality criteria for population structure information. We added real data about real people to the dataset where we were confident the people who hadn’t completed a census form (which is known as admin enumeration) will be counted. We also used data from the 2018 and 2013 Censuses, administrative data sources, and statistical imputation methods to fill in some missing characteristics of people and dwellings.

​

Data quality

The quality of data in the 2023 Census is assessed using the quality rating scale and the quality assurance framework to determine whether data is fit for purpose and suitable for release. Data quality assurance in the 2023 Census has more information.

​

Concept descriptions and quality ratings

Data quality ratings for 2023 Census variables has additional details about variables found within totals by topic, for example, definitions and data quality.

​

Using data for good

Stats NZ expects that, when working with census data, it is done so with a positive purpose, as outlined in the Māori Data Governance Model (Data Iwi Leaders Group, 2023). This model states that "data should support transformative outcomes and should uplift and strengthen our relationships with each other and with our environments. The avoidance of harm is the minimum expectation for data use. Māori data should also contribute to iwi and hapū tino rangatiratanga”.

​

Confidentiality

The 2023 Census confidentiality rules have been applied to 2013, 2018, and 2023 data. These rules protect the confidentiality of individuals, families, households, dwellings, and undertakings in 2023 Census data. Counts are calculated using fixed random rounding to base 3 (FRR3) and suppression of ‘sensitive’ counts less than six, where tables report multiple geographic variables and/or small populations. Individual figures may not always sum to stated totals. Applying confidentiality rules to 2023 Census data and summary of changes since 2018 and 2013 Censuses has more information about 2023 Census confidentiality rules.

​

Measures

Measures like averages, medians, and other quantiles are calculated from unrounded counts, with input noise added to or subtracted from each contributing value during measures calculations. Averages and medians based on less than six units (e.g. individuals, dwellings, households, families, or extended families) are suppressed. This suppression threshold changes for other quantiles. Where the cells have been suppressed, a placeholder value has been used.

​

Percentages

To calculate percentages, divide the figure for the category of interest by the figure for 'Total stated' where this applies.

​

Symbol

-997 Not available

-999 Confidential

​

Inconsistencies in definitions

Please note that there may be differences in definitions between census classifications and those used for other data collections.

According to a study conducted in 2024 using the most recently available data, the average poverty rate in news deserts in the United States (counties without access to or with very limited access to local news) was around five percent higher than the country average, at ** percent. Citizens living in counties without newspapers were also earning a lower median annual income than the general population average, with the figure estimated at less than ****** U.S. dollars compared to more than **** thousand U.S. dollars for the U.S. as a whole.

Graph and download economic data for Real Median Household Income in Michigan (MEHOINUSMIA672N) from 1984 to 2023 about MI, households, median, income, real, and USA.

VITAL SIGNS INDICATOR List Rents (EC9)

FULL MEASURE NAME List Rents

LAST UPDATED October 2016

DESCRIPTION List rent refers to the advertised rents for available rental housing and serves as a measure of housing costs for new households moving into a neighborhood, city, county or region.

DATA SOURCE real Answers (1994 – 2015) no link

Zillow Metro Median Listing Price All Homes (2010-2016) http://www.zillow.com/research/data/

CONTACT INFORMATION vitalsigns.info@mtc.ca.gov

METHODOLOGY NOTES (across all datasets for this indicator) List rents data reflects median rent prices advertised for available apartments rather than median rent payments; more information is available in the indicator definition above. Regional and local geographies rely on data collected by real Answers, a research organization and database publisher specializing in the multifamily housing market. real Answers focuses on collecting longitudinal data for individual rental properties through quarterly surveys. For the Bay Area, their database is comprised of properties with 40 to 3,000+ housing units. Median list prices most likely have an upward bias due to the exclusion of smaller properties. The bias may be most extreme in geographies where large rental properties represent a small portion of the overall rental market. A map of the individual properties surveyed is included in the Local Focus section.

Individual properties surveyed provided lower- and upper-bound ranges for the various types of housing available (studio, 1 bedroom, 2 bedroom, etc.). Median lower- and upper-bound prices are determined across all housing types for the regional and county geographies. The median list price represented in Vital Signs is the average of the median lower- and upper-bound prices for the region and counties. Median upper-bound prices are determined across all housing types for the city geographies. The median list price represented in Vital Signs is the median upper-bound price for cities. For simplicity, only the mean list rent is displayed for the individual properties. The metro areas geography rely upon Zillow data, which is the median price for rentals listed through www.zillow.com during the month. Like the real Answers data, Zillow's median list prices most likely have an upward bias since small properties are underrepresented in Zillow's listings. The metro area data for the Bay Area cannot be compared to the regional Bay Area data. Due to afore mentioned data limitations, this data is suitable for analyzing the change in list rents over time but not necessarily comparisons of absolute list rents. Metro area boundaries reflects today’s metro area definitions by county for consistency, rather than historical metro area boundaries.

Due to the limited number of rental properties surveyed, city-level data is unavailable for Atherton, Belvedere, Brisbane, Calistoga, Clayton, Cloverdale, Cotati, Fairfax, Half Moon Bay, Healdsburg, Hillsborough, Los Altos Hills, Monte Sereno, Moranga, Oakley, Orinda, Portola Valley, Rio Vista, Ross, San Anselmo, San Carlos, Saratoga, Sebastopol, Windsor, Woodside, and Yountville.

Inflation-adjusted data are presented to illustrate how rents have grown relative to overall price increases; that said, the use of the Consumer Price Index does create some challenges given the fact that housing represents a major chunk of consumer goods bundle used to calculate CPI. This reflects a methodological tradeoff between precision and accuracy and is a common concern when working with any commodity that is a major component of CPI itself. Percent change in inflation-adjusted median is calculated with respect to the median price from the fourth quarter or December of the base year.

Korea Average: AH: Less Than 30: Savings data was reported at 19,880.000 KRW th in 2017. This records a decrease from the previous number of 20,910.000 KRW th for 2016. Korea Average: AH: Less Than 30: Savings data is updated yearly, averaging 20,395.000 KRW th from Mar 2010 (Median) to 2017, with 8 observations. The data reached an all-time high of 23,520.000 KRW th in 2015 and a record low of 15,890.000 KRW th in 2011. Korea Average: AH: Less Than 30: Savings data remains active status in CEIC and is reported by Statistics Korea. The data is categorized under Global Database’s Korea – Table KR.H078: SHFLC: Household Assets, Liabilities & Income By Age Groups of Households Head (10Age).

BackgroundZambia is among the 30 high-burden countries for tuberculosis (TB), Human Immunodeficiency Virus (HIV)-associated TB, and multi-drug resistant/rifampicin resistant TB with over 5000 children developing TB every year. However, at least 32% of the estimated children remain undiagnosed. We assessed healthcare workers’ (HCWs) knowledge, attitudes, and practices (KAP) towards childhood TB and the factors associated with good KAP towards childhood TB.MethodsData was collected at two primary healthcare facilities in Lusaka, Zambia from July to August 2020. Structured questionnaires were administered to HCWs that were selected through stratified random sampling. Descriptive analysis was done to determine KAP. A maximum knowledge, attitude, and practice scores for a participant were 44, 10, and 8 points respectively. The categorization as either “poor” or “good” KAP was determined based on the mean/ median. Logistic regression analysis was performed to assess the associations between participant characteristics and KAP at statistically significant level of 0.05%.ResultsAmong the 237 respondents, majority were under 30 years old (63.7%) and were female (72.6%). Half of the participants (50.6%) were from the outpatient department (OPD) and antiretroviral therapy (ART) clinic, 109 (46.0) had been working at the facility for less than 1 year, 134 (56.5%) reported no previous training in TB. The median/mean KAP scores were 28 (IQR 24.0–31.0), 7 (IQR = 6.0–8.0) and 5 points (SD = 1.9) respectively. Of the participants, 43.5% (103/237) had good knowledge, 48.1% (114/237) had a good attitude, and 54.4% (129/237) had good practice scores on childhood TB. In the multivariate analysis, clinical officers and individuals with 1–5 years’ work experience at the facility had higher odds, 2.61 (95% CI = 1.18–5.80, p = 0.018) and 3.09 (95% CI = 1.69–5.65, p = 0.001) of having good attitude respectively, and medical doctors had 0.17 lower odds (95% CI = 0.18–5.80, p = 0.018) of good childhood TB practice. Other participant characteristics didn’t show a significant association with the scores.ConclusionThe study found suboptimal levels of knowledge, attitude, and practices regarding childhood TB among HCWs. Targeted programmatic support needs to be provided to address the above gaps.

In 2023, public housing residents in Alaska, Arkansas, and the U.S. Virgin Islands had the highest household incomes compared to their respective local median incomes in the United States. In these areas, the average public housing household incomes constituted at least ** percent of the local median income. In contrast, states like Maryland, Ohio, Washington, Guam, and the District of Columbia exhibited the lowest proportions, where households housed in social housing earned less than ** percent of the local median income.

Bangladesh HIES: Average Monthly Income: Less than BDT 3,000 data was reported at 995.370 BDT in 2022. Bangladesh HIES: Average Monthly Income: Less than BDT 3,000 data is updated yearly, averaging 995.370 BDT from Dec 2022 (Median) to 2022, with 1 observations. The data reached an all-time high of 995.370 BDT in 2022 and a record low of 995.370 BDT in 2022. Bangladesh HIES: Average Monthly Income: Less than BDT 3,000 data remains active status in CEIC and is reported by Bangladesh Bureau of Statistics. The data is categorized under Global Database’s Bangladesh – Table BD.H010: Household Income and Expenditure Survey: Average Monthly Income per Household: by Income Group.