PIRLS 2021 Results
SCALE CREATION
The PIRLS achievement scales were developed using a combination of item response theory and latent regression models and are reported using multiple imputations in the form of five plausible values. Each participating student receives five plausible values in the reported scales.
The PIRLS achievement scales range from 5 to 995, with a center point of 500 and a standard deviation of 100 based on the pooled sample of participating countries in 2001.
To measure trends over time, the results of all subsequent PIRLS assessments (2006, 2011, 2016, and 2021) are transformed to this metric.
To ensure comparability of the results over time, results for paperPIRLS and digitalPIRLS were jointly scaled in PIRLS 2021 as part of a bridge study. A linear transformation was used to put the PIRLS 2021 paperPIRLS and digitalPIRLS results on the 2016 PIRLS reported metric.
LIST OF ACHIEVEMENT SCALES
- Overall achievement (Reading)
- Purposes for Reading
- Literary Experience
- Acquire and Use Information
- Processes of Comprehension
- Retrieving and Straightforward Inferencing
- Interpreting, Integrating, and Evaluating
Overall achievement results were also reported in terms of the percentages of students reaching the following four international benchmarks:
- Advanced International Benchmark (625)
- High International Benchmark (550)
- Intermediate International Benchmark (475)
- Low International Benchmark (400)
A scale-anchoring process was used to describe the competencies of students who met each of these standards.
SCALE CREATION
To report the results of the context questionnaire, PIRLS 2021 developed a number of background questionnaire scales. All scales were constructed using item response theory (IRT), specifically the partial credit model.
For certain scales with common items across previous cycles of PIRLS, the scale scores were transformed to the existing reporting metric.
In addition, scale scores were categorized into categorical variables by classifying respondents into one of three groups for each scale: High, Middle, and Low. The cut-off points on the scale delimiting the regions were described in terms of combinations of response categories. These criteria are described in the Methods and Procedures: PIRLS 2021 Technical Report.
LIST OF BACKGROUND SCALES
- Home environment support
- Home Socioeconomic Status
- Home Resources for Learning
- Early Literacy Activities
- Early Literacy Tasks
- Early Literacy and Numeracy Activities
- Early Numeracy Activities
- Parents Like Reading
- School climate
- School Emphasis on Academic Success
- School Emphasis on Academic Success – Teachers’ Reports
- School Emphasis on Academic Success – Principals’ Reports
- Sense of School Belonging
- Teachers’ Job Satisfaction
- School discipline and safety
- School Discipline
- Safe and Orderly School
- Student Bullying
- Students’ reading attitudes and behaviors
- Students Like Reading
- Students Confident in Reading
- Students Engaged in Reading Lessons
- Students’ beliefs about their familiarity with digital devices
- Students’ Digital Self-Efficacy
- Classroom instruction
- Parents’ Perception of Their Child’ Classroom
- Disorderly Behavior During Reading Lessons
- Teaching Limited by Student Not Ready for Instruction
- Classroom Teaching Limited by Student Not Ready for Instruction
- Instruction Affected by Reading Resource Shortages
Reading achievement
Results regarding average reading achievement and reading achievement distributions for the 43 countries and 5 benchmarking participants that assessed grade 4 students at the end of the school year:
- They had relatively high achievement in reading, i.e., fourth-grade students in almost three-fourths (30 out of the 43 countries) had higher achievement than the scale centerpoint of 500.
- The achievement distributions show a large within-country range in many of PIRLS 2021 countries (about 200 points or even larger between lower- and higher-performing students).
Achievement and gender
- Fourth-grade girls had higher average achievement than boys in almost all the countries, with an average advantage of 16 points across the 43 countries and 18 points across the 57 countries.
- The difference in achievement between boys and girls was relatively small in Spain, the Czech Republic, Israel, Malta, and Iran.
Performance at the international benchmarks
- Advanced International Benchmark
- Students reaching this level interpreted, integrated, and evaluated story plots and information in relatively complex text.
- Only small percentages of students reached this level. In the 43 countries with data collection at the end of grade 4:
- Half of the countries had 7 percent or fewer students reaching the advanced level.
- Six countries had 15% or more of their students reach this benchmark: Singapore had by far the highest amount (35%) of its fourth-grade students reaching the advanced level. Then, after a gap, Hong Kong SAR and the Russian Federation each had 21 percent followed by England (18%), Bulgaria (16%), and Sweden (15%).
- Low International Benchmark
- Most of the PIRLS 2021 countries were able to educate 94 percent of their fourth-grade students to a basic level (Low Benchmark), this is close to universal basic literacy in the majority of the PIRLS 2021 countries.
Trend
- Comparison with the last cycle (2016-2021)
- Out of the 43 countries that collected their PIRLS 2021 data at the end of the fourth grade, 32 countries (and one benchmarking participant) had also participated in 2016. These are their trends:
- 21 countries (and the 1 benchmarking entity) had lower average reading achievement in 2021.
- 8 countries had no or little change in achievement.
- 3 had higher average achievement in 2021.
- Out of the 43 countries that collected their PIRLS 2021 data at the end of the fourth grade, 32 countries (and one benchmarking participant) had also participated in 2016. These are their trends:
- Trends over 20 years (2001-2021)
- 15 countries had comparable data across four or five assessments since 2001:
- In general, they have had their “ups and downs”.
- Singapore was the only country that demonstrated a steady improvement in five PIRLS assessments.
- 18 countries participated in the 2001 and 2021 cycles:
- 7 countries showed progress.
- 6 countries had a similar achievement.
- 5 countries showed decline.
- 15 countries had comparable data across four or five assessments since 2001:
- Average achievement by gender (for the countries with trend data from previous PIRLS assessments)
- Overall, little progress has been made in closing the reading achievement gender gap favoring girls.
- Although 21 out of the countries with data collection in grade 4 had lower average achievement in 2021 than in 2016, for the most part the decreases in achievement were similar for girls and boys. There was little narrowing (or widening) in the gender gap favoring girls.
Home
- There was a strong correlation between students' socioeconomic status and their academic success:
- On average, 29% of students were classified as having “higher”, 48% “middle”, and 23% “lower” home SES.
- The results show a large difference of 86 points in average reading achievement between students with “higher” SES and “lower” SES (542 vs. 456).
- Across countries, higher average reading achievement was related to:
- Parents engaging their children in early literacy activities before primary school
- Parents who like to read
- Children’s literacy readiness when beginning primary school
School
Across the countries, students had higher reading achievement on average if they attended schools with:
- Higher proportions of economically affluent than disadvantaged students
- Greater percentages of students beginning primary school equipped with basic literacy skills
- Where instruction was not affected by reading resource shortages
- A higher degree of emphasis on academic success
- A higher degree of discipline and safety
Students’ reading attitudes and behaviors
- Liking reading
- In general, fourth-grade students had positive attitudes about reading: “very much like reading” (42% of students on average), “somewhat like reading” (40%) and “do not like reading” (18%)
- There was a modest relationship between liking reading and reading achievement. Students who responded that they “do not like reading” had the lowest average reading achievement.
- Students had an accurate self-assessment of their own level of reading.
- Students use of digital devices to find and read information:
- Three categories of time spent were identified: “more than 30 minutes per school day” (25% of students), “30 minutes or less per school day” (52%) and “no time per school day” (23%)
- There was a correlation between the time students spent on digital devices and their achievement, i.e., students who reported spending
- “30 minutes or less per school day” had the highest average reading achievement
- “more than 30 minutes per school day” had somewhat lower achievement
- “no time” had the lowest average reading achievement.