With information from both parents and students, the TIMSS 2019 results add to the already extensive body of research showing a powerful positive relationship between students’ socioeconomic environment and their educational achievement.

The Home Resources for Learning scale combines data reported by fourth grade students and their parents. The parents’ data were collected using the TIMSS 2019 Early Learning Survey (also called the “Home Questionnaire”), which asked parents to provide information about their child’s learning experiences before the child began primary school. As explained in Exhibit 5.1, students provided information about the number of books in their homes and other study supports, while parents provided information about the number of children’s books, the parents’ levels of education, and the parents’ occupations (see About the Scale). As also explained, students were assigned a score on the scale according to the availability of these five home resources for learning.

Exhibit 5.2 presents, for each country, the percentage of students in three categories of the scale, together with average student achievement in mathematics. Countries are ordered by the percentage of students in the “many resources” category, with an average percentage across countries of 17 percent. The majority of students (75%, on average) were assigned to the “some resources” category, with just 8 percent in the “few resources” category. There was a 129-point difference in average mathematics achievement of students in the “many resources” category compared with the “few resources” category (562 vs. 433). Average mathematics achievement for the students in the “some resources” category was in between, at 498 points.

The second panel of Exhibit 5.2 presents a scatterplot showing a modest positive association between countries’ average Home Resources for Learning and average mathematics achievement.

Exhibit 5.3 presents the same Home Resources for Learning data, but this time in conjunction with average science achievement. The overall pattern of results is similar to mathematics, with an average science achievement gap of 143 score points between students in the “many resources” and “few resources” categories (557 vs. 414). Average science achievement for the students in the “some resources” category was again in between, at 488 points. Also similar to mathematics, the scatterplot in the second panel of Exhibit 5.3 shows a modest positive relationship between average Home Resources for Learning scores and average science achievement across countries.

The TIMSS 2019 results reinforce the already extensive body of research showing a powerful positive relationship between students’ socioeconomic environment and their educational achievement. As shown in Exhibit 5.4, the eighth grade Home Educational Resources scale is based on students’ reports about the number of books and other study supports in their homes, as well as the highest level of their parents’ education (see About the Scale). As also explained, students were assigned a score on the scale according to the availability of these three home educational resources.

Exhibit 5.5 presents for each country the percentage of students in three categories of the scale, together with average student achievement in mathematics. Countries are ordered by the percentage of students in the “many resources” category, with an average percentage across countries of 14 percent. Educational resource availability varied widely across countries, with the percentage of students in the “many resources” category ranging from 2 percent in South Africa and Morocco to 40 percent in the Korea. The majority of students (73%, on average) were assigned to the “some resources” category, with 13 percent in the “few resources” category. There was a 113-point difference in the average mathematics achievement of students in the “many resources” category compared with the “few resources” category (546 vs. 433). Average mathematics achievement for the students in the “some resources” category was in between, at 488 points.

The second panel of Exhibit 5.5 presents a scatterplot showing the relationship between average Home Educational Resources and average mathematics achievement across countries. There is a modest positive association between Home Educational Resources and average mathematics achievement at the country level.

Exhibit 5.6 presents the same Home Educational Resources data, but this time in conjunction with average science achievement. The overall pattern of results is similar to mathematics, with an average science achievement gap of 118 score points between students in the “many resources” and “few resources” categories (549 vs. 431). Average science achievement for the students in the “some resources” category was again in between, at 489 points. Also similar to mathematics, the scatterplot in the second panel of Exhibit 5.6 shows a modest positive relationship between average Home Educational Resources scores and average science achievement across countries.

Exhibits 5.7 and 5.8 show the percentages of fourth grade students that “always,” “almost always,” “sometimes,” or “never” speak the language of the TIMSS test at home, together with average achievement in mathematics (Exhibit 5.7) and science (Exhibit 5.8). Exhibits 5.9 and 5.10 provide comparable information for eighth grade students.

About two-thirds of students (63% at fourth grade and 66% at eighth grade), on average, reported “always” speaking the language of the test at home, and most of the remaining students (32% at fourth grade and 30% at eighth grade) reported speaking it “almost always” or “sometimes.” There was relatively little variation in average achievement across these categories. However, the few students (5% at fourth grade and 4% at eighth grade, on average) who “never” spoke the language of the test at home had much lower average achievement in both mathematics (442 for both fourth and eighth grades) and science (418 and 428, respectively).

Considerable research, including results from TIMSS and PIRLS, has documented the importance of early childhood learning activities for fostering student achievement during later school years. The Early Literacy and Numeracy Activities scale focuses specifically on literacy and numeracy activities, summarizing parents’ reports about how often they engaged with their child in nine literacy activities and nine numeracy activities before their child began primary school (see About the Scale in Exhibit 5.11).

Exhibit 5.12 presents the percentage of students whose parents reported “often,” “sometimes,” and “never or almost never” engaging them in early literacy and numeracy activities, together with average fourth grade mathematics achievement. Countries are ordered by the percentage of students whose parents “often” engaged them in early literacy and numeracy activities. In general, parents reported a good deal of early interaction with their children, with 42 percent of the students having parents who “often” engaged them in such activities and an additional 55 percent with parents who “sometimes” engaged them, on average. Just 3 percent of students had parents who reported “never or almost never” engaging them in early literacy or numeracy activities.

There was a modest positive relationship between the frequency of engagement reported and later mathematics achievement, with the students in the “often” category having higher average achievement than students in the “sometimes” category (516 vs. 495, respectively). In several countries, a small percentage of students (3%, on average) had parents who “never or almost never” engaged them in early literacy and numeracy activities, and these students typically had lower average mathematics achievement (456).

Exhibit 5.13 shows the same student percentages for the Early Literacy and Numeracy Activities scale as Exhibit 5.12, but this time together with average fourth grade science achievement. The activities-science achievement relationship was similar to the activities-mathematics achievement relationship, with highest achievement for those whose parents “often” engaged them (507), next for those who were “sometimes” engaged (484), and lowest for those who were “never or almost never” engaged (421).

There was a positive relationship for fourth grade students between the number of years they attended preprimary education and their mathematics achievement (Exhibit 5.14) and science achievement (Exhibit 5.15). According to their parents, a majority of fourth grade students across countries had attended three years or more of preprimary school—56 percent. Beyond that, on average, 17 percent had attended two years, 15 percent attended one year or less, and 12 percent did not attend preprimary school.

The students who attended three years or more had the highest average mathematics achievement (509), and those with less preprimary school attendance had successively lower average achievement (495, 483, and 464, respectively). There was a similar pattern with regard to science achievement, with an average of 500 for those who attended three years or more, and successively lower science achievement for those with less preprimary attendance (489, 472, and 452, respectively).

Exhibits 5.16 and 5.17 provide further insight into the relative effects of formal preprimary education and informal literacy and numeracy activities in the home on later achievement in mathematics and science, respectively. The results suggest that preprimary education cannot completely replace parental involvement in preparing children for later success in school. At each level of preprimary attendance, from three years or more down through one year or less, the students whose parents “often” engaged them in early literacy and numeracy activities had higher average mathematics and higher average science achievement than those engaged only “sometimes” or “never.”

For both mathematics and science, average achievement was highest (520 and 513, respectively) among the 26 percent of students that had three years or more of preprimary education and whose parents “often” engaged them in literacy and numeracy activities before beginning primary school. It was lowest (469 and 455, respectively) among the 18 percent that had one year or less of preprimary education and parents reporting only “sometimes” or “never” engaging them in literacy and numeracy activities. See About the Scale in Exhibit 5.11 for information about the Early Literacy and Numeracy Activities scale.

To provide information about the extent to which students enter primary school equipped with some basic literacy and numeracy skills as a foundation for learning, the TIMSS Early Learning Survey (or “Home Questionnaire”) included a set of questions asking parents how well their child could do various literacy and numeracy activities when he or she first began primary school. Twelve items formed the Literacy and Numeracy Tasks scale in TIMSS 2019 (see About the Scale in Exhibit 5.18).

Parents’ reports indicate that early preparation appears to have an effect on mathematics and science achievement through the fourth grade. Exhibit 5.19 shows that, on average across countries, 25 percent of students entered school able to perform early literacy and numeracy tasks “very well” according to their parents, and more than half (51%) entered school able to perform the tasks “moderately well.” Parents’ assessments of their children’s early literacy and numeracy skills corresponded well with mathematics achievement at the fourth grade, with the children able to perform “very well” having higher achievement than those performing “moderately well” (532 vs. 498). The 24 percent of the students in the “not well” category had the lowest average achievement (468). Exhibit 5.20 provides a similar picture in relation to science achievement. Students in the “very well” category had the highest average science achievement (518) compared with students in the “moderately well” and “not well” categories (488 and 461, respectively).

As shown in Exhibit 6.1, TIMSS characterizes schools in terms of socioeconomic composition based on principals’ reports of the percentages of economically disadvantaged and economically affluent students in the school (see About the Scale). “More affluent” schools were defined as having more than 25 percent of students from economically affluent homes and not more than 25 percent from economically disadvantaged homes, while “more disadvantaged” schools had more than 25 percent of students from disadvantaged homes and not more than 25 percent from affluent homes. All other combinations were considered to be “neither more affluent nor more disadvantaged.”

Exhibits 6.2 and 6.3 present the percentages of fourth grade students in each category in relation to mathematics achievement and science achievement, respectively. At fourth grade, 41 percent of students, on average, were in schools with relatively more affluent students than disadvantaged students, 34 percent were in schools where there was a balance, and 25 percent were in schools with relatively more disadvantaged than affluent students. Average achievement was highest for students in the “more affluent” schools for both mathematics (521) and science (512) and lowest in the “more disadvantaged” schools (479 and 467 for mathematics and science, respectively). Average achievement for students attending “neither more affluent nor more disadvantaged” schools was in the middle (499 for mathematics and 489 for science).

Exhibits 6.4 and 6.5 present the percentages of eighth grade students in each category in relation to mathematics achievement and science achievement, respectively. Compared with the fourth grade, the distribution of eighth grade students was more even across the school composition groups, with 35 percent in schools with relatively more affluent students than disadvantaged students, 33 percent in schools where there was a balance, and 32 percent in schools with relatively more disadvantaged students than affluent students. The relationship with achievement was similar to fourth grade, with average achievement highest for students in “more affluent” schools for both mathematics (518) and science (518) and lowest in the “more disadvantaged” schools (466 and 468 for mathematics and science, respectively). Average achievement for students attending “neither more affluent nor more disadvantaged” schools was in the middle (489 for mathematics and 490 for science).

Principals’ categorizations of their schools according to the percentage of fourth grade students who spoke the language of the TIMSS 2019 assessment as their first language are shown in Exhibit 6.6, together with average mathematics achievement, and in Exhibit 6.7 together with science achievement. On average, 63 percent were in schools where most students (more than 90%) spoke the language of the TIMSS assessment as their first language, and another 18 percent were in schools where more than half of students (51–90%) spoke the language of the assessment as their first language. Both groups of students had higher average achievement than the 19 percent of students attending schools where only half of the students (or less) spoke the language of the assessment as their native language (506 and 501 vs. 486 for mathematics, and 498 and 493 vs. 471 for science).

As shown in Exhibit 6.8 for mathematics and in Exhibit 6.9 for science, the distribution of eighth grade students across the language composition groups was similar to fourth grade. Sixty-four percent of students were in schools where most students spoke the assessment language as their first language (more than 90%) and 17 percent and 18 percent were in schools where more than half of the students (51–90%) and only half of the students (or less) did so, respectively. There was little difference in average mathematics or science achievement among the groups, although average science achievement was lower in schools where only half of the students (or less) spoke the language of the assessment as their native language (483 vs. 484 and 487 for mathematics, and 479 vs. 491 and 490 for science).

To collect information about the ability composition of the student intake into the primary grades, TIMSS asked school principals to estimate the percentage of students in the school who possess a range of literacy and numeracy skills when beginning primary school (see About the Scale in Exhibit 6.10).

Exhibits 6.11 and 6.12 present the percentages of students in each of three groups of schools categorized by their principals’ reports, together with average mathematics and science achievement, respectively. There was considerable variation across countries in the percentage of students in the highest category—“more than 75% enter with skills”—from 89 percent in Ireland, where many students start preprimary school after their fourth birthday (as described in the TIMSS 2019 Encyclopedia), to 0 percent in the Czech Republic and Hungary. On average, 24 percent of students were in schools where “more than 75%” of the students entered school with literacy and numeracy skills, 56 percent were in schools where “25–75%” entered with such skills, and 20 percent in schools where “less than 25%” enter with literacy and numeracy skills.

For both mathematics and science, average achievement was higher for students in schools with greater percentages of entering students having literacy and numeracy skills. Students attending schools where “more than 75% enter with skills” had the highest achievement, followed by schools where “25–75% enter with skills,” and students in schools with “less than 25%” having the lowest average achievement (508, 499, and 480, respectively, for mathematics and 499, 490, and 471, respectively, for science).

The Instruction Affected by Mathematics Resource Shortages scale, described in Exhibit 6.13, summarizes principals’ reports about two kinds of resource shortages affecting instruction: general school resources and resources specific to mathematics instruction (see About the Scale). In Exhibit 6.14, countries are ordered (from most to least) according to the percentage of fourth grade students in schools “not affected” by resource shortages, from a high of 68 percent in Korea to a low of 0 percent in Kosovo. On average, 26 percent of fourth grade students attended schools “not affected” by resource shortages, and they had the highest average mathematics achievement (514). Sixty-eight percent of the students were in schools “somewhat affected” by resource shortages, and 6 percent were in schools “affected a lot.” Average mathematics achievement for these two groups was 499 and 473, respectively.

Exhibit 6.15 presents the results of the Instruction Affected by Mathematics Resource Shortages scale for eighth grade students. Similar to fourth grade, the percentage of students in schools “not affected” by resource shortages ranged from a high of 73 percent to a low of 2 percent, with an average of 30 percent. Sixty-three percent of the students were in schools “somewhat affected” by resource shortages, and 7 percent were in schools “affected a lot.” As in fourth grade, average mathematics achievement was highest for students in schools “not affected” by resource shortages (509), next for students in “somewhat affected” schools (483), and lowest for students in schools that were “affected a lot” (476).

The Instruction Affected by Science Resource Shortages scale, like its counterpart for mathematics, includes principals’ reports about two kinds of resource shortages affecting instruction: general school resources and resources specific to science instruction (see About the Scale in Exhibit 6.16). On average, 24 percent of fourth grade students were in schools “not affected” by science resource shortages, 69 percent were in schools “somewhat affected,” and 7 percent were in schools “affected a lot” (see Exhibit 6.17). Average science achievement ranged across the three categories from a high of 508 for students in schools “not affected” by resource shortages, to 488 for students in schools “somewhat affected,” to a low of 472 for students in schools “affected a lot.”

Eighth grade students’ results for the Instruction Affected by Science Resource Shortages scale are shown in Exhibit 6.18. On average, 30 percent of students were in schools “not affected” by science resource shortages, 62 percent were in schools “somewhat affected,” and 8 percent were in schools “affected a lot.” Similar to the mathematics scale, there was a modest negative association between the degree students’ schools were affected by resource shortages and average science achievement, with achievement highest for students in schools “not affected” by resource shortages (510), next highest in “somewhat affected” schools (484), and lowest in schools that were “affected a lot” (472).

The main education paths for becoming a school principal can vary across countries (as described in the TIMSS 2019 Encyclopedia). Some countries focus on formal education or completion of specialized training and certification, while others focus on years of teaching experience and gradual promotion into school leadership roles.

As shown in Exhibit 6.19, on average, more than half of fourth grade students (54%) had a school principal with a postgraduate university degree, 42 percent had a principal with a bachelor’s degree, and 5 percent had a principal who had not completed a bachelor’s degree. On average, 68 percent of fourth grade students had a principal with an educational leadership certificate or license, and 36 percent with a postgraduate leadership degree.

At the eighth grade (see Exhibit 6.20), the situation with regard to school principals’ formal education was rather similar, with 55 percent of students having a school principal with a postgraduate university degree, 43 percent having a principal with a bachelor’s degree, and 2 percent with a principal who had not completed a bachelor’s degree. On average, 71 percent of eighth grade students had a principal with an educational leadership certificate or license, and 39 percent had a principal with a postgraduate leadership degree.

Exhibit 6.21 presents fourth grade students’ school principals’ reports about their years of experience. On average, fourth grade students’ principals had 10 years of experience as a principal. Thirteen percent of the students had very experienced principals with 20 years or more of experience, 29 percent had principals with at least 10 years of experience, 27 percent had principals with at least 5 years of experience, and 31 percent had principals with less than 5 years of experience.

Exhibit 6.22 shows eighth grade students’ school principals’ reports about their years of experience. On average, principals of schools with eighth grade students had 9 years of experience as a principal. Eleven percent of the students had principals with 20 years or more of experience, 28 percent had principals with at least 10 years of experience, 28 percent had principals with at least 5 years of experience, and 34 percent had principals with less than 5 years of experience.

The School Emphasis on Academic Success scale was administered to school principals of fourth and eighth grade students to collect information about the school’s expectations toward academic achievement. As described in Exhibit 7.1, students were categorized on the scale according to their principals’ responses to 11 items in the TIMSS 2019 School Questionnaire (see About the Scale). Consistent with previous TIMSS results, principals overall had positive attitudes about the degree of emphasis on academics in their schools, so the three regions were described as “very high,” “high,” and “medium.”

Exhibits 7.2 and 7.3 present the results for fourth grade students. The two exhibits include the same results for the School Emphasis on Academic Success scale, but Exhibit 7.2 presents results in relation to mathematics achievement and Exhibit 7.3 in relation to science achievement. Countries are ordered according to the percentage of students in the “very high emphasis” category. Across countries, on average, 7 percent of fourth grade students attended schools where the principal reported a “very high emphasis” on academic success, 55 percent attended schools with a “high emphasis,” and 37 percent attended schools with a “medium emphasis.” Attending schools with a higher emphasis on academic success was related to higher average mathematics achievement. Students in the “very high emphasis” category had the highest average mathematics achievement (515), followed by the “high emphasis” category (508), and then students in the “medium emphasis” category (486). The results were similar in relation to science, with average science achievement for students in the three categories at 508, 499, and 474 points, respectively.

Exhibits 7.4 and 7.5 present the School Emphasis on Academic Success results for eighth grade students in relation to their mathematics achievement and science achievement, respectively. On average, the principals of eighth grade students reported a similar degree of emphasis on academic success compared with fourth grade—8 percent of eighth grade students attended schools with a “very high emphasis” on academic success, 49 percent attended schools with a “high emphasis,” and 43 percent attended schools with a “medium emphasis.” At the eighth grade, there was a greater achievement gap between the “very high emphasis” and “medium emphasis” categories than at the fourth grade (69 points vs. 29 points in mathematics, and 68 points vs. 34 points in science).

The TIMSS 2019 Early Learning Survey (or “Home Questionnaire”) asked the parents of fourth grade students about the extent to which they are satisfied that their child’s school promotes academic standards and fosters a positive school climate. Students were categorized on the Parents’ Perceptions of Their Child’s School scale according to their parents’ responses to eight items listed in Exhibit 7.6 (see About the Scale).

Exhibit 7.7 presents the results for the Parents’ Perceptions of Their Child’s School in relationship to students’ mathematics achievement. Overall, parents reported positive perceptions about their child’s school. On average across countries, the majority of fourth grade students (64%) had parents who reported being “very satisfied,” but these percentages ranged from 90 percent in Albania to 15 percent in Korea and 9 percent in Japan. On average, 31 percent of students attended schools with which their parents were “somewhat satisfied.” Only 5 percent of students on average had parents who were “less than satisfied” with their child’s school. Average mathematics achievement was similar across the three categories of the scale, with students in the “very satisfied” category having only slightly higher achievement than students in the “somewhat satisfied” and “less than satisfied” categories (504 vs. 497 and 495, respectively).

Exhibit 7.8 shows the same results for fourth grade students, but in relationship to science achievement. On average, students had very similar science achievement across the three scale categories (493, 487, and 491, respectively).

The items forming the Sense of School Belonging scale asked students about their attitudes toward school, including the extent to which they like being in school, feel that they belong, and have good relationships with teachers. Both fourth and eighth grade students were asked how much they agreed or disagreed with the five statements listed in Exhibit 7.9 (see About the Scale).

The percentages of fourth grade students assigned to the three categories on the TIMSS 2019 Sense of School Belonging scale are shown in Exhibit 7.10 in relation to mathematics achievement and in Exhibit 7.11 in relation to science achievement. Overall, the majority of fourth grade students had positive attitudes toward school, and each successive category of increased school belonging was related to an increase in average achievement. More than half of students (58%) reported having a “high sense of school belonging,” and these students had the highest average mathematics achievement (508) and average science achievement (497). On average, 34 percent of students had a “moderate sense of school belonging” with comparatively lower achievement (498 in mathematics and 487 in science). Only 8 percent of fourth grade students had “little sense of school belonging,” and these students had the lowest average achievement (484 in mathematics and 476 in science).

Exhibits 7.12 and 7.13 present the same results for eighth grade students in relation to mathematics achievement and science achievement, respectively. Compared with the fourth grade, smaller percentages of eighth grade students reported having a “high sense of school belonging,” and larger percentages reported having a “moderate sense of school belonging.” On average, across countries, 37 percent of eighth grade students reported a “high sense of school belonging,” and 49 percent had a “moderate sense of school belonging.” Like the fourth grade, there was a positive association between eighth grade students’ sense of school belonging and their average achievement in mathematics and science. The 14 percent of students with “little sense of school belonging” had the lowest average achievement (470 in both mathematics and science). Students in the “high sense of school belonging” and “moderate sense of school belonging” categories had comparatively higher average achievement (500 and 489, respectively, in mathematics, and 502 and 490, respectively, in science).

Continuing from previous TIMSS assessments, TIMSS 2019 asked school principals for their perceptions about the extent that discipline, disorder, and bullying behaviors are problems in their school. As described in Exhibit 8.1, fourth grade students were assigned scores on the School Discipline scale according to their principals’ characterizations of 10 discipline and safety problems (see About the Scale). The eighth grade scale included one additional problem more suited to older students—“physical injury to teachers or staff.” Since TIMSS first reported results on the School Discipline scale in 2011, data have shown that attending schools with fewer discipline and safety problems was associated with higher average achievement in mathematics and science.

Exhibit 8.2 presents the results for the School Discipline scale at the fourth grade in relation to mathematics achievement. Countries are ordered according to the percentage of students in schools where principals reported “hardly any problems” with discipline and safety. On average, the majority of fourth grade students (60%) attended schools with “hardly any problems,” and 32 percent attended schools with “minor problems.” Only 8 percent of students, on average, attended schools for which principals reported “moderate to severe problems” with discipline and safety. Consistent with previous TIMSS results, average mathematics achievement was higher for students in schools with “hardly any problems” than for students in schools with “minor problems” (508 vs. 494). At 466 points, on average, mathematics achievement for students in schools with “moderate to severe problems” with school discipline and safety was substantially lower (42 points) than for students in schools with “hardly any problems.”

Exhibit 8.3 presents the same percentages of fourth grade students in the three categories of the scale, but in relation to science achievement. Similar to the results for mathematics, fewer discipline problems were related to higher average science achievement. Students in schools with “hardly any problems” with discipline and safety had the highest average science achievement (498), followed by students in schools with “minor problems” (483) and “moderate to severe problems” (457).

Exhibits 8.4 and 8.5 present the School Discipline results for eighth grade students in relation to mathematics achievement and science achievement, respectively. Compared with the fourth grade, fewer eighth grade students attended schools with “hardly any problems” (45% vs. 60%) and more eighth grade students attended schools with “minor problems” (43% vs. 32%). Similar to fourth grade results, the 11 percent of eighth grade students in the “moderate to severe problems” category had the lowest average achievement (448 and 452 in mathematics and science, respectively). However, eighth grade results showed a greater difference in average achievement between students in schools with “hardly any problems” and those in schools with “moderate to severe problems.” There was a 55-point gap in average mathematics achievement between the two categories (503 vs. 448) and a 52-point gap in average science achievement (504 vs. 452).

In addition to principals’ reports, previous TIMSS assessments have also asked teachers about their perceptions of school safety and found that a strong perception of safety and order in school was related to higher average achievement in mathematics and science. For TIMSS 2019, teachers of students in both the fourth and eighth grade assessments were asked the degree to which they agreed or disagreed with eight statements on the Safe and Orderly School scale (see About the Scale in Exhibit 8.6). Based on their teachers’ responses, students were categorized as being in “very safe and orderly” schools, “somewhat safe and orderly” schools, and “less than safe and orderly” schools.

Exhibits 8.7 and 8.8 present the percentages of fourth grade students in the three categories in relation to average mathematics achievement and average science achievement, respectively. Across countries, on average, more than half the fourth grade students (61%) attended schools judged by their teachers to be “very safe and orderly,” with almost all of the remaining students (36%) in schools perceived to be “somewhat safe and orderly.” Internationally, on average, students attending “very safe and orderly” schools as reported by their teachers had the highest average mathematics achievement (507) and science achievement (497) compared with students in “somewhat safe and orderly” schools (495 in mathematics and 484 in science). Only small percentages of students (4%, on average) were in schools judged to be “less than safe and orderly.”

Exhibits 8.9 and 8.10 present the corresponding Safe and Orderly School scale results at the eighth grade in relation to mathematics achievement and science achievement, respectively. Similar to the fourth grade, almost all the eighth grade students (93% on average across countries) were in “very safe and orderly” or “somewhat safe and orderly” schools, but smaller percentages of eighth grade students were in the “very safe and orderly” category, and larger percentages of were in the “somewhat safe and orderly” category. On average, 48 percent of eighth grade students attended schools judged by their teachers to be “very safe and orderly,” 45 percent were in “somewhat safe and orderly” schools, and 6 percent were in “less than safe and orderly” schools. There was a strong positive association between eighth grade students’ average achievement and their teachers’ reports about school safety, with students in “very safe and orderly” schools having average mathematics achievement 41 points higher than students in “less than safe and orderly” schools (501 vs. 460). Similarly, students in “very safe and orderly” schools had average science achievement 35 points higher than students in “less than safe and orderly” schools (501 vs. 466).

Since 2011, TIMSS has asked students about how often they experienced various bullying behaviors by their school peers, such as being teased, excluded from activities, or physically hurt. The TIMSS 2019 fourth grade Student Bullying scale is described in Exhibit 8.11 (see About the Scale). Students were assigned to one of three categories of the scale (“never or almost never,” “about monthly,” “about weekly”) according to their reports of how often they experienced 11 bullying behaviors.

Exhibits 8.12 and 8.13 present the scale results in relation to mathematics achievement and science achievement, respectively. Internationally, the majority of fourth grade students (63%, on average) reported to be “never or almost never” bullied by school peers, 29 percent reported being bullied “about monthly,” and only 8 percent were bullied “about weekly.” Reports from fourth grade students’ about being bullied were related to their average achievement in mathematics and science. Across countries, there was a negative association between fourth grade students’ reports about the frequency of being bullied and average achievement, with students bullied “about weekly” having substantially lower average mathematics achievement (451) and lower average science achievement (437) than students bullied “about monthly” (495 and 486 in mathematics and science, respectively). Students in the “never or almost never” category had the highest achievement on average—512 in mathematics and 503 in science.

The eighth grade Student Bullying scale used in previous TIMSS assessments was revised in TIMSS 2019 to better reflect current trends related to social media and cyberbullying. As described in Exhibit 8.14, the scale in 2019 included 14 items with an emphasis on bullying experienced through digital devices (see About the Scale). Students were assigned to one of three scale categories (“never or almost never,” “about monthly,” “about weekly”) according to their reports of how often they experienced these 14 behaviors.

Exhibits 8.15 and 8.16 provide the results for the TIMSS 2019 eighth grade Student Bullying scale in relation to mathematics achievement and science achievement, respectively. On average, more than two-thirds of eighth grade students (71%) reported being bullied “never or almost never,” and these students had the highest average achievement—496 in mathematics and 499 in science. With each reported increase in the frequency they experienced bullying, eighth grade students had progressively lower average achievement to the extent that there was a 68-point difference in mathematics achievement and a 78-point difference in science achievement between students in the “never or almost never” category and students in the “about weekly” category. Students bullied “about monthly” also had much higher achievement than students bullied “about weekly” (482 vs. 428 in mathematics and 482 vs. 421 in science).