Quality notes
Age
The methods used to produce the 2011 Census estimates were designed to achieve the highest quality at fiveyear age groups. This priority was agreed with the key users of the data. This means that estimates by single year of age will be of lower quality than those for fiveyear age groups because they will be subject to greater statistical errors because of sampling variance and imputation effects.
The sampling error associated with the 2011 Census estimates is mainly dependent on the Census Coverage Survey (CCS) sample size, the size of the population, the census response rate, the CCS response rate and the degree of similarity of the population the error level relates to. At fiveyear age group level, the overall error will be smaller than the error associated with single year of age estimates, however sample sizes do vary between age groups and therefore some error levels may be smaller or larger than average. Read more about sampling errors in Confidence intervals for the 2011 Census.
Tenure
The census asked respondents to identify who their landlord was and the results reflect the responses they gave. In the past decade, some local authorities in Wales have transferred the management of all their local authority housing stock to other social landlords. Individuals responding to the census reported their understanding of their landlord and this may not reflect the actual management arrangements in all cases.
The number of households reporting that they were renting from the local authority may be too high while those reporting that they were renting social housing may be too low. This amount cannot be quantified. Changes between 2001 and 2011 need to be interpreted in this context.
Statistics on social and private rented tenure types can be found in tables:

KS402EW  2011 Census: Tenure

QS403EW  2011 Census: Tenure  People

QS404EW  2011 Census: Tenure Household Reference Person aged 65 and over

QS405EW  2011 Census: Tenure – Households
Clarifications
Calculation of percentages in the Key Statistics tables
In all Key Statistics tables, the percentage calculation in each cell uses the overall population of the table as the denominator, even where the table contains a subpopulation breakdown. For example, where a column shows a breakdown of figures by age for males, the percentages for each age group will use the total population as the denominator and not the total population of males.
Calculation of rooms and bedrooms required for occupancy rating
Occupancy rating provides a measure of whether a household's accommodation is overcrowded or under occupied. There are two measures of occupancy rating, one based on the total number of rooms in a household's accommodation, and one based only on the number of bedrooms.
This explains how we calculate the number of bedrooms and rooms required:
The total room requirements for a multiperson household are:

One room per couple or lone parent.

One room per person aged 16 and above who is not a lone parent or in a couple.

One room for every two males aged 1015, rounded down.

One room for every pair of males of whom one is aged 1015 and one is aged 09, if there are an odd number of males aged 1015.

One room for a remaining unpaired male aged 1015 if there are no males aged 09 to pair him with.

Repeat steps 35 for females.

One room for every two remaining children aged 09 (regardless of gender), rounded up.

Add two rooms to this total.
The bedroom requirements for a multiperson household are:

One bedroom per couple.

One bedroom per person aged 21 or over not in a couple.

One bedroom for every two males aged 1020, rounded down.

One bedroom for every pair of males of whom one is aged 1020 and one is aged 09, if there are an odd number of males aged 1020.

One bedroom for a remaining unpaired male aged 1020 if there are no males aged 09 to pair him with.

Repeat steps 35 for females.

One bedroom for every two remaining children aged 09 (regardless of gender), rounded up.
Example 1
Based on the requirements, a oneperson household requires three rooms in total. The same household requires one bedroom. This is based on step 1 and 8 for rooms and step 1 for bedrooms.
Example 2
This example is based on a household with two parents with male children aged 17, 13, and 8, and two female children both aged 11:

One room for the two parents. [Running total =1]

One room for the 17yearold. [Running total =2]

There is one male aged 1015, so this count is zero (0.5 is rounded down). [Running total =]

One room for the 13yearold paired with the 8yearold. [Running total =3]

There are no remaining unpaired males aged 1015. [Running total =3]

One room for the two female 11yearolds. [Running total =4]

There are no remaining children aged 09. [Running total =4]

Add two. [Total =6]
This household requires six rooms in total.

One bedroom for the two parents. [Running total =1]

No other people aged 21 and over are in the household. [Running total =1]

There are two males aged 1020 so one bedroom is needed. [Running total =2]

There is an even number of males aged 1020, so this step does not apply.

Nor does this one.

There are two female 11yearolds so one bedroom is needed. [Running total =3]

There is one 8yearold so one bedroom is needed. [Total =4]
This household requires four bedrooms.
These tables contain information about occupancy rating:
KS403EW, QS408EW, QS412EW
Travel to work
In 2001, if a respondent recorded that their workplace address was at home, it was assumed that they did not travel for work, and they were recorded as working at home in the travel to work data.
In 2011, this assumption has not been made. This means that any respondent who works at home but also reported a method of travel to work (e.g. a mobile hairdresser) is recorded as working at home but is also recorded as using whichever mode of travel they provided in the travel to work data. This is a more accurate reflection of the population characteristics and is more useful and relevant to transport planners and other users.
This means that the 2001 table KS015EW is not directly comparable with the 2011 table (QS701EW). To allow comparison with 2001, ONS has released an additional table CT0015 which gives an alternative take on the 2011 data at local authority level, and is compatible with 2001 Census data.