Photo 1: Image showing a school in Sierra Leone. Photo credit: Chris McBurnie (EdTechHub)

In Sierra Leone, the Ministry of Basic and Senior Secondary Education (MBSSE) and the Teaching Service Commission (TSC) are developing plans to increase the equity and efficiency of teacher allocation. 

Over the past few years, the government has struggled to retain qualified teachers in remote and rural locations. Today, the pupil-to-qualified-teacher ratio has risen from 44:1 for schools in urban areas to 76:1 for schools in remote areas (see this 2020 research and policy paper by Mackintosh and colleagues). 

To address these challenges, the MBSSE and the TSC have started exploring  innovative ways, such as using a teacher preference matching model. EdTech Hub, the Education Commission, and Fab Inc are undertaking research to support this work and understand the factors influencing teacher mobility, including when and how teachers choose a new school.

In Sierra Leone, a government payroll position is tied to the teacher, not the school, meaning teachers can change schools once they go on the payroll. This leads to certain schools being chronically understaffed. It also suggests that deploying teachers to rural and understaffed schools (or putting teachers already there on payroll) might not tackle high pupil-to-qualified-teacher ratios in the longer term. Investigating the school-to-school movement of payroll teachers can help us understand where teachers go after they go on the payroll and how long they stay in a given school. We also looked at overall movement nationally to identify areas of the country that are outliers in terms of high and low retention of teachers.

How did we find patterns?

Using the annual school censuses (ASC) from 2015–2021 (excluding 2016, due to lack of information on key variables for this study) we connected school data to individual teachers. The ASC is completed by school leaders each year and contains data about a range of issues, including location (including GPS coordinates) and teachers’ background details (age, sex, payroll numbers, position, qualifications, subject speciality, and source of payment, among other things). Using the ASC data, we looked at teachers’ movements in four ways: through movement analysis, spatial analysis, benchmarking findings, and retention analysis. We did this nationally at primary and secondary levels. 

  1. Movement analysis

Based on the ASC, government payroll teachers were identified and assigned a unique code. These were then mapped against their school codes for 2015–2021, allowing us to follow teachers’ movements across schools for six consecutive years. In addition, we looked to see if teachers move schools the year after going on the government payroll. The TSC has considered using payroll status as an incentive to teach in rural schools. As mentioned above, payroll status is tied to the teacher, not the school, so it is important to understand if teachers stay in their schools after being put on the government payroll. The analysis excluded payroll teachers from 2015 and earlier, as the ASC data does not tell us when a teacher went on the payroll for those years. 

  1. Spatial analysis

In addition to matching teachers to school data, we conducted a spatial analysis using the GRID3 classification of settlement areas. This classification system uses building density to categorise remoteness. Specifically, we looked at movements between hamlets, small settlements, and built-up areas, defined as follows: 

Settlement type Characteristics
Built-up areas (BUAs)• Areas of urbanisation
• Visible grid of streets and blocks
• Polygons that maintain 100m2 building density of 13 or more across an area ⪰0.4 km2
Small settlements (SSAs)• Areas of permanently inhabited structures and compounds of roughly a few hundred to a few thousand inhabitants
• Assemblage of family compounds adjoining other similar habitations
• Polygons containing 50 or more buildings, but not a BUA
Hamlets (HAM)• Collection of several compounds or sleeping houses in isolation from SSAs or BUAs
• Polygons containing between 1 and 49 buildings
(Source: GRID3, 2021)

For any movement, we have a source and a destination school and their coordinates. This allows us to estimate the distance between the schools and observe the direction of teacher movement.

  1. Benchmarking findings

While we can describe the patterns in the data, drawing conclusions about what is ‘normal’ — or a common pattern of movement — is difficult because we do not have other movement data for Sierra Leone or similar contexts to compare against. That is why we benchmarked findings against the random movement of teachers. We are testing the null hypothesis that when a teacher chooses to move from their current school, they have no preference for which other school they move to except that the other school is willing to let them teach there. This allows us to find patterns that were influenced by factors other than simply the availability of a position at another school. 

  1. Retention analysis

Finally, we investigated which schools had high retention rates (i.e. schools where many teachers remain employed) and which schools had low retention rates (i.e. schools where many teachers leave their position). We looked at yearly retention (proportion of teachers who remain at a given school for two consecutive years) between 2015–2017, 2017–2018, 2018–2019, 2019–2020, 2020–2021 and six-year retention rates (proportion of teachers who remained at a given school for all the years between 2015 and 2021). Through this retention analysis, we identified ‘hot spots’ (schools with high retention rates) and ‘cold spots’ (schools with low retention rates). 

So where do teachers go, and where do teachers stay?

When do teachers move

We found that 25% of all teachers in our sample moved at some point. We also found that female teachers are more likely to move than male teachers. Mobility in the year after having been put on the payroll is low. However, 8.1 % of male and 9.6 % of female teachers still move in that year. Overall, teachers without qualifications are less likely to move. At primary level, teachers with a bachelor’s degree are the most likely to move. At secondary level, teachers with a postgraduate degree are the most likely to move schools. 

How far do teachers move?

When teachers move schools, they move 18.6 km away on average. Looking more closely, we find that 56.2% of all teachers move less than 5 km away. Compared to the random movement of teachers, we found that for all distances under 42 km, teachers move more often. Teachers move less often for all distances greater than 50 km, and for distances between 42 and 50 km, there is no significant difference. This shows that teachers move more often than we would expect for distances shorter than 42 km. Secondly, we found that teachers move more frequently within settlement types (i.e. BUA, SSA, Hamlets) and that most movements occur in BUAs, as shown in the table below.

When looking at movement at the school level, we found that primary school teachers moved more between the same settlement types. In contrast, secondary school teachers had more mixed preferences. When moving locally (i.e. within 5 km), they had a preference for moving within the same settlement; but for moving distances greater than 5 km, no such patterns hold. When looking at the complete sample, we do not see any difference in movement between male and female teachers. However, female teachers move shorter distances at the pre-primary and primary levels, while female teachers at the secondary level move further away. 

Where do teachers stay?

Retention rates by school level

In addition to patterns of individuals’ movements, we also looked at school-level retention rates of payroll teachers. We found that the year-to-year retention rates of our overall sample are, on average, 84%, with primary having the highest retention (85%) and senior secondary having the lowest (74%). However, we saw significantly lower retention when we looked at the six-year retention rates. Nearly half of the primary and two-thirds of the secondary-level payroll teacher workforce left their posts between 2015 and 2021. 

Retention rates by school location

Across all schools, we found that the further schools are from the district headquarters, the higher the year-to-year retention rates. The six-year retention rates showed no significant difference. The table below shows the retention rates per school type.

School typeYear-to-year retentionSix-year retention
OverallThe further away from the district headquarters, the higher the retention ratesNo difference
Primary levelThe further away from the district headquarters, the higher the retention ratesNo difference
Secondary levelLower retention rates for distances greater than 50 km from district headquartersLower retention rates for distances shorter than 5 km from district headquarters

Retention rates by settlement type

Across all schools, BUAs had the lowest retention rates, both for year-to-year retention and six-year retention, and hamlets had the highest. For the secondary level, there were lower retention rates in SSAs than BUAs, but similar retention rates in BUAs and hamlets. 

Retention rates by administrative area

Finally, we also looked at differences in retention rates across administrative areas, as seen in the table below.

Highest retention ratesLowest retention rates
year-to-yearsix-yearyear-to-yearsix-year
PrimaryFalabaKailahunFalaba PujehunWestern Area Urban
Koinadugu
Western Area Urban
Koinadugu
SecondaryFalaba
Bombali
Western Area UrbanFalabaKenemaKono
Moyamba
KonoBonthe
Conclusions

Based on these findings, we can draw the following conclusions:

  1. Our findings show that teachers’ movements are localised: teachers move shorter distance ranges and within settlement types. This suggests that the labour market is localised. 
  2. Overall, female teachers move more frequently than male teachers and move further at the secondary level. Given the discrepancy in the proportion of female teachers in the workforce at primary and secondary levels (ASC 2021), this warrants further investigation.
  3. The overall average retention rate per year is 84%, which is much lower at the secondary level. Overall BUAs have the lowest retention rates. Investigating why certain schools have high and low retention rates is needed to better understand teachers’ motivations for moving schools. 
  4. Movement patterns for teachers in primary and secondary schools are distinct and need to be investigated separately. 

Next steps: why do teachers move schools?

To find out why teachers are leaving particular schools and why some schools have high retention rates, we identified a list of ‘hotspots’ and ‘coldspots’ at the school level, organised by district, settlement type, and school level (primary, secondary). Given that female teachers are more likely to move at primary level and further at the secondary level, particularly in light of discrepancies in the workforce, the movements of female teachers in particular need further investigation. In the next phase of our research, we will conduct intensive qualitative fieldwork consisting of interviews and focus group discussions with teachers and school leaders in hot and cold spots. As the current study identified clear differences between primary and secondary levels, we will include this distinction in our data collection and analysis.

Do you want to learn more about teacher mobility patterns and retention in Sierra Leone?

Connect with Us

Get a regular round-up of the latest in clear evidence, better decisions, and more learning in EdTech.

Connect with Us​

Get a regular round-up of the latest in clear evidence, better decisions, and more learning in EdTech.

EdTech Hub is supported by

The findings, interpretations, and conclusions expressed in the content on this site do not necessarily reflect the views of The UK government, Bill & Melinda Gates foundation or the World Bank, the Executive Directors of the World Bank, or the governments they represent.

EDTECH HUB 2024. Creative Commons Attribution 4.0 International License.

to top