Are some school inspectors more lenient than others?

Figures & data

Table 1. Descriptive statistics for the distribution of inspector characteristics.

	Primary		Secondary
	Short	Not short	Short	Not short
Lead inspector contract
Her Majesty’s Inspector	59%	20%	60%	45%
Ofsted Inspector	41%	80%	40%	55%
Lead inspector gender
Female	54%	48%	44%	43%
Male	45%	51%	55%	56%
unknown	0%	0%	0%	0%
Primary/secondary specialism
Primary inspections only	65%	73%	0%	0%
70%–99% primary	19%	17%	7%	13%
30%–69% primary	14%	9%	38%	40%
Secondary inspections only	1%	1%	55%	47%
Inspection outside home region
Yes	2%	16%	3%	15%
No	90%	67%	83%	62%
Not available	7%	17%	14%	23%
Academic year
2011/12	0%	19%	0%	15%
2012/13	0%	23%	0%	21%
2013/14	0%	19%	0%	17%
2014/15	0%	15%	0%	14%
2015/16	12%	6%	18%	7%
2016/17	33%	5%	36%	7%
2017/18	34%	6%	32%	9%
2018/19	21%	7%	15%	9%
Previous inspections led
M	33	29	19	19
SD	30	28	17	20
minimum	1	1	1	1
25^th percentile	11	8	6	5
50^th percentile	25	20	14	12
75^th percentile	43	42	26	26
maximum	186	182	103	161
Team size
1 inspector	83%	28%	12%	7%
2 inspectors	6%	35%	56%	7%
3 inspectors	6%	33%	8%	26%
4 inspectors	4%	4%	9%	44%
5	1%	0%	15%	15%
n	8,329	21,521	1,199	4,747

Table 2. Cross-tabulations between characteristics of the lead inspector and Overall Effectiveness judgements.

(a) Gender
	Primary			Secondary
	Female	Male	Difference	Female	Male	Difference
Outstanding	7.8	8.2	0.4	10.9	10.1	−0.9
Good	55.9	58.7	2.9	45.4	44.9	−0.5
Requires Improvement	30.5	28.6	−1.9	34.6	34.6	−0.1
Inadequate	5.9	4.5	−1.4	9.1	10.5	1.4
n (inspections)	11,056	11,698		2,188	2,813
(b) Contract status
	Primary			Secondary
	OI	HMI	Difference	OI	HMI	Difference
Outstanding	7.7	9.0	−1.3	10.5	10.3	−0.2
Good	60.3	47.0	−13.3	47.8	42.2	−5.6
Requires Improvement	27.8	35.4	7.7	33.9	35.4	1.6
Inadequate	4.2	8.6	4.4	7.8	12.1	4.3
n (inspections)	17,622	5,139		2,654	2,370
(c) Inspection outside of home region
	Primary			Secondary
	No	Yes	Difference	No	Yes	Difference
Outstanding	8.4	7.4	−1.0	9.1	14.7	5.6
Good	56.8	58.9	2.1	44.6	44.5	−0.1
Requires Improvement	29.5	28.5	−1.0	35.7	33.6	−2.1
Inadequate	5.2	5.2	0.0	10.6	7.2	−3.4
n (inspections)	15,925	3,347		3,161	735
(d) Primary/secondary specialism
	Primary			Secondary
	Split	Primary only		Split	Second only
Outstanding	11.0	7.2	−3.8	10.2	11.1	0.9
Good	52.9	58.2	5.3	43.9	46.1	2.2
Requires Improvement	30.1	29.5	−0.6	35.3	33.1	−2.2
Inadequate	6.0	5.0	−1.0	10.6	9.8	−0.8
n (inspections)	1,912	15,871		1,976	2,308

Note: OI = Ofsted inspector; HMI = Her Majesty’s Inspector.

Table 3. Differences in inspection assignments by gender and contract status of the lead inspector (primary schools).

	Gender		Her Majesty’s Inspector
	Female	Male	No	Yes
Inspection type
Section 5	68%	70%	74%	49%
Requires Improvement reinspection	20%	18%	18%	22%
Academy first Section 5	5%	4%	3%	10%
Section 8 deemed Section 5	4%	4%	4%	7%
Serious weakness inspection	1%	1%	1%	4%
Exempt school inspection	2%	2%	0%	8%
Section 8 no formal designation	0%	0%	0%	1%
Missing	0%	0%	0%	0%
Prior inspection rating
Outstanding	8%	8%	7%	13%
Good	41%	40%	42%	37%
Requires Improvement	43%	45%	47%	32%
Inadequate	4%	4%	2%	14%
Missing	3%	3%	3%	5%
Free school meals (FSM) quintile
Quintile 1 (Low FSM)	16%	17%	17%	15%
Quintile 2	19%	20%	20%	18%
Quintile 3	21%	21%	21%	20%
Quintile 4	22%	22%	21%	24%
Quintile 5 (High FSM)	22%	20%	20%	23%
Missing	0%	0%	0%	0%
School absence quintile
Quintile 1 (low absences)	20%	21%	21%	19%
Quintile 2	23%	22%	23%	21%
Quintile 3	23%	23%	23%	24%
Quintile 4	21%	21%	21%	22%
Quintile 5 (high absences)	13%	13%	13%	14%
Missing	0%	0%	0%	0%
Key Stage 2 English quintile
Quintile 1 (low achievement)	24%	24%	23%	31%
Quintile 2	21%	20%	21%	20%
Quintile 3	17%	17%	18%	14%
Quintile 4	17%	16%	17%	14%
Quintile 5 (high achievement)	12%	13%	13%	9%
Missing	9%	10%	10%	10%
Key Stage 2 maths quintile
Quintile 1 (low achievement)	24%	23%	22%	31%
Quintile 2	19%	20%	20%	18%
Quintile 3	20%	18%	19%	20%
Quintile 4	15%	15%	16%	12%
Quintile 5 (high achievement)	12%	13%	13%	10%
Missing	9%	10%	10%	10%

Table 4. Ordinal logistic regression model results of the association between lead inspector characteristics and Overall Effectiveness grades.

	Primary		Secondary
	OR	t	OR	t
Gender (Ref: Male)
Female	0.84*	−3.18	1.09	1.16
Contract status (Ref: Ofsted Inspector)
Her Majesty’s Inspector	1.45*	6.21	1.32*	3.49
Experience (Ref: Bottom quintile)
Quintile 2	0.94	−0.6	0.72	−1.95
Quintile 3	1.00	0.03	0.71*	−2.13
Quintile 4	1.07	0.69	0.82	−1.08
Quintile 5 (most experienced)	0.93	−0.64	0.89	−0.53
Inspection outside home region (Ref: No)
Yes	1.13*	2.50	0.99	−0.14
Phase specialism (ref: primary/secondary only)
30%–69% primary	0.86	−1.91	0.92	−1.06
70%–99% primary	0.95	−0.83	1.03	0.21
Team size
1 inspector	Reference		0.43*	−4.92
2 inspectors	1.25*	5.29	0.90	−0.93
3 inspectors	1.26*	5.23	1.08	0.93
4 inspectors	1.05	0.56	Reference
5 + inspectors	0.68	−1.60	0.83*	−2.10

Note: Data based upon inspections conducted between the 2011/12 to 2018/19 academic years. Estimates based on 22,754 inspections conducted by 983 inspectors. Standard errors have been clustered at the inspector level. Models control for percent of pupils eligible for free school meals, school religion, school gender, region, inspection type, prior Ofsted rating, school performance data, school absences, percent of pupils at the school with special educational needs, percent of pupils who speak English as an additional language, and other inspector characteristics.

*p < .05.

Table 5. Logistic regression model estimates of the association between lead inspector characteristics and primary school short inspection outcomes.

	n	OR	t
Gender (Ref: Male)
Female	3,605	0.75*	2.17
Contract (Ref: Ofsted inspector)
Her Majesty’s Inspector	3,627	1.63*	3.56
Experience (Ref: Bottom quintile)
Quintile 2	3,627	0.65	−1.85
Quintile 3		0.84	−0.73
Quintile 4		1.05	0.22
Quintile 5 (most experienced)		1.22	0.92
Inspection outside home region (Ref: No)
Yes	3,311	1.41	1.08
Phase specialism (ref: primary only)
30%–69% primary	3,594	1.41	1.64
70%–99% primary		1.36	1.77
Team size (Ref: 1 inspector)
2 inspectors	3,511	0.99	−0.05

Note: Sample restricted to short primary school inspections conducted between January 2018 and August 2019. Dependent variable coded 1 if the short inspection resulted in the outcome “Section 5 next due to concerns” or was immediately converted to a full inspection resulting in a judgement of Inadequate or Requires Improvement, and zero otherwise. Models have been estimated separately for each characteristic. Models control for percent of pupils eligible for free school meals, school religion, school gender, region, inspection type, prior Ofsted rating, school performance data, school absences, percent of pupils at the school with special educational needs, percent of pupils who speak English as an additional language, and other inspector characteristics.

*p < .05.

Table 6. Cross-tabulation between inspection team size and inspection outcomes.

(a) Primary
	Team size
	1	2	3	4	5+
Overall effectiveness
Outstanding	9	7	8	14	26
Good	61	59	55	46	30
Requires Improvement	27	29	31	33	41
Inadequate	3	5	7	8	3
N	5,546	7,184	7,158	1,093	150
(b) Secondary
	Team size
	1	2	3	4	5+
Overall effectiveness
Outstanding	22	9	7	10	15
Good	43	47	46	44	46
Requires Improvement	30	36	35	35	32
Inadequate	4	9	12	11	7
N	233	273	1,148	2,072	889

Note: Figures refer to column percentages.

Table 7. Predicted distribution of primary school inspection outcomes for two hypothetical inspectors. Ordinal logistic regression model estimates.

	Inspector A	Inspector B	Risk ratio (A/B)
Overall effectiveness
Outstanding	4.5%	9.0%	0.50
Good	48.0%	59.3%	0.81
Requires Improvement	38.4%	27.2%	1.41
Inadequate	9.1%	4.5%	2.03
Short inspection
Conversion with downgrade or Section 5 next due to concerns. (Jan18–Aug19)	15.5%	9.7%	1.60
Inspector characteristics
Team size	2 inspectors	1 inspector
Contract status	Her Majesty’s Inspector	Ofsted Inspector
Gender	Female	Male

Note: Model controls for percent of pupils eligible for free school meals, region, previous Ofsted inspection outcome, inspection type, Key Stage 2 maths and English scores, school absences, percent of pupils with English as an additional language, whether the inspection was conducted after 2018, school religion, school gender composition, Key Stage 1 scores, and percent of pupils with special educational needs.