Foundation Stage Profile (Age 5)

Each pupil is assessed by their class teacher at the end of their first year of school (at the end of reception) which coincides with the end of the early years foundation stage (EYFS). This assessment is called the early years foundation stage profile (EYFSP), but was called the Foundation Stage Profile in 2006/2007.

The FSP replaced baseline assessment in the reception year in 2002/2003 (QCDA). Between 2002/2003 and 2005/2006 NPD contains FSP for only 10% of pupils, however. This is because Local Education Authorities (LEAs) were only required to submit a 10% sample of complete individual pupil FSP data (National Strategies Briefing for Headteachers). LEAs have submitted complete individual pupil FSP data since 2006/2007 (SFR).

The primary purpose of the Early Years Foundation Stage Profile is provide year 1 teachers with reliable and accurate information about each child's level of development as they reach the end of the EYFS (EYFSP Handbook).

The EYFS profile sums up and describes each child’s development and learning achievements at the end of the EYFS. It is based on ongoing observation and assessment by the class teacher in six areas of learning and development. These are:
  • personal, social and emotional development
  • communication, language and literacy
  • problem solving, reasoning and numeracy
  • knowledge and understanding of the world
  • physical development
  • creative development

These six areas of learning are formed from 13 scales. Each scale is from 0-9, where scale point 0 describes a child for whom it has not been possible to record an assessment because of their individual needs, points 1-3 describe a child working towards the achievements described in the Early Learning Goals, points 4-8 are drawn from the Early Learning Goals themselves and point 9 describes a child who has developed further than the level of the Early Learning Goals (FSP User Guide).

The scale name and variable name in NPD are given in the table below:
Area of learning
Scale
Variable name in NPD
Personal, social and emotional development
Dispositions and attitudes
pse_as1

Social development
pse_as2

Emotional development
pse_as3
Communication, language and literacy
Language for communication and thinking
cll_as1

Linking sounds and letters
cll_as2

Reading
cll_as3

Writing
cll_as4
Problem solving, reasoning and numeracy
Numbers as labels and for counting
mat_as1

Calculating
mat_as2

Shape, space and measures
mat_as3
Knowledge and understanding of the world
Knowledge and understanding of the world
kuw_aol
Physical development
Physical development
phy_aol
Creative development
Creative development
cre_aol

NPD also provides total scores for personal, social and emotional development (pse_aol), communication, language and literacy (cll_aol) and problem solving, reasoning and numeracy (mat_aol) and a total score for EYFSP (fsp_tot).
For detailed information about each EYFSP scale see chapter 7 of the EYFSP Handbook.

Data collection

EYFSP is assessed and reported by teachers. Schools report data to DfE via the TeacherNet School 2 School website. Schools should report the thirteen EYFSP scale summary scores for each child. For each scale a score between 0 and 9 or ‘N’ should be reported. ‘N’ (not assessed) applies if it was not possible to assess the child because of late arrival in the school or absence due to prolonged sickness (Essex Schools Info Link).

Validity of measure

The EYFSP is assessed and reported by teachers. The EYFSP Handbook says that these judgements should be made “through assessing behaviour that a child demonstrates consistently and independently in a range of situations” and also that “practitioners should initially consider each scale point separately”. For each scale point, the judgement made should represent assessment of the child’s typical attainment.

These teacher assessments should also be moderated in each school.

Recent research by IFS used EYFSP as a proxy for early attainment in literacy to find suitable comparison pupils for those in other schools receiving an early literacy intervention. The CLL scale was most appropriate, but it was found that CLL was not a good predictor for those that received the intervention.

Cleaning the variables

Stata code:
The variables in NPD are relatively clean, although they are stored as string variables. For most purposes they must be converted to numeric (rather than string) format. An example of this for the total score is below:
destring fsp_tot, gen(fsp_tot) force
This code will replace all non-numeric characteristics to missing. This means that those pupils that are given scale point 'N' are recoded to missing. This will generally be appropriate but should be noted.

A total score is still reported for areas of learning and development where one or more component is missing. For example, in 2006/2007 total score was reported for communication, language and literacy although 92 pupils were missing cll_as1, 252 pupils were missing cll_as2, 52 pupils were missing cll_as3 and 180 pupils were missing cll_as4. Researchers may therefore want to use a mean score for each component based on non-missing scales. The following code generates a mean score for the communication, language and literacy area of learning and development:
foreach var of varlist cll_as1 cll_as2 cll_as3 cll_as4 cll_aol {
destring `var', replace force
}
egen rownonmiss=rownonmiss(cll_as1 cll_as2 cll_as3 cll_as4)
gen cll_mean=cll_aol/rownonmiss
label var cll_mean "Mean score across all non-missing CLL components"

The following code generates a binary variable for having a mean CLL score in the lowest 25% of all pupils in the cohort:
cap drop clllow25
sum cll_mean, de
gen clllow25=.
replace clllow25=0 if cll_mean>`r(p25)' & cll_mean<.
replace clllow25=1 if cll_mean<=`r(p25)'

SPSS code
Checks needed

Description of values across cohorts


The table below shows the FSP total score over time, split by pupils that are and are not eligible for FSM.

Non FSM pupils
FSM pupils
Total
2002/2003
90.1
79.4
88.3
2003/2004
92.0
81.4
90.2
2004/2005
90.5
78.7
88.4
2005/2006
87.6
75.6
85.6
2006/2007
87.3
76.1
85.4
2007/2008



2008/2009



Analysis by the QCDA suggests that from 2006 to 2009 developments in the way assessment and moderation have been conducted have contributed to the historic downturn in the percentage of children achieving at the higher end of the scales (8 or 9 points). These decreases in the number of children are levelling off in 2010 and may suggest that the reporting system may now have become embedded (SRF).