5 Results
5.2 Applying language environment analysis on Finnish twin data: what does
5.2.1 Do speaker segment durations in LENA analyses differ between
5.2.1.1 Key child segment duration
Group comparison with monthly inspections of key child segment durations and neonatal health-related variables did not reveal constant significant differences between any of the groups (see Appendix 2 for statistical information). In addition, no single statistically significant group difference were found for key child segment durations of birth weight, birth complications, or the level of neonatal care in any points of measure.
However, preterm and full-term twins’ daily segment durations were found to differ statistically significantly for key child segment durations at 10 and 11 month recordings.
Preterm twins vocalized 25.2% less than term twins at 10 and 26.8% less than full-term twins at 11 months of CA.
When comparing key child segment durations from the pooled data of groups of twins differing in neonatal health-related background variables, no statistically significant differences were found between LBW and NBW twins, between preterm and full-term twins, or between twins born with and without birth complications (Table 10).
Statistically significant differences were found in twin groups differing in the level of neonatal care; twins cared at a neonatal ward were on average 9.5% more voluble than twins cared at bedside, but the difference showed only a small practical significance.
Table 10. Information on central decencies, dispersion, and statistical differences in group comparisons of the effect of neonatal background variables to key child volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin preterm 42.48 10.74 40.84 13.41 55 2024.00 -1.54 0.12 0.15 Twin full-term 39.96 11.25 38.16 13.99 87
Twin LBW 42.26 13.14 40.71 15.99 68 2246.00 -1.10 0.27 0.11
Twin NBW 39.72 8.69 38.13 11.84 74
No birth
complications 39.82 10.95 38.16 14.21 93 1932.00 -1.49 0.14 0.15 Born with
complications 43.05 11.13 40.84 11.24 49
Cared at bedside 39.35 9.51 38.03 13.67 95 1743.00 -2.12 0.03 0.22 Cared at a ward 44.13 13.27 42.47 12.25 47
No constant statistically significant group differences were found for key child segment durations in monthly group comparisons of demographic background variables, although two single statistically significant differences were detected
(Appendix 3). In addition, key child durations were not found to differ for twins differing in mother and father education, nor between twin As and Bs in any points of measure.
Key child gender showed, however, a statistically significant group difference at 7 months and having an older sibling in 10-month measure points. Boy twins were on average 21.2% less voluble at seven-month recordings, when compared with twin girls and twins with siblings, who were on average 21.5% less voluble than twins without older siblings.
Table 11 presents statistical information about group comparisons of the pooled data for demographic background variables of key child segment duration. It was found that female twins’ daily overall key child segment durations exceeded durations of key child male twins by 9.8%, but the effect size from the difference remained small. In addition, the mother’s education did show a significant group difference with a small effect size for key child segment durations. Mothers with higher degrees had twin children who were on average 10.2% more voluble than twins of mother’s with lower education. In addition, in families with no older siblings, key children were on average 46.9% more voluble than key children living in families with older siblings. However, the effect size of the mother’s education on child volubility also remained small. This is explained by the proportion of overlapping variance of the observed key child durations, which is visualized in Figure 6. From other background variables, birth order and the father’s education did not yield statistically significant differences in the overall daily durations of key children.
Table 11. Information on central decencies, dispersion, and statistical differences in group comparisons of the effect of demographic background variables to key child volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin A 41.15 12.09 40.51 14.71 71 2511.0 -0.04 0.97 0.00
Twin B 40.72 10.06 38.27 11.44 71
Twin girl 42.58 11.91 41.88 12.47 86 1906.0 -2.10 0.04 0.21 Twin boy 38.41 9.22 37.83 10.86 56
No older siblings 73.07 12.03 40.66 12.88 71 2024.0 -2.03 0.04 0.20 Older siblings 38.80 9.67 7.70 13.56 71
Mom lower
education 38.82 9.43 37.53 11.46 74
2012.0 -2.06 0.04 0.20 Mom higher
education 43.23 12.30 40.58 16.26 68 Dad lower
education 41.00 10.17 38.44 11.64 72
2495.0 -1.01 0.31 0.01 Dad higher
education 40.86 12.02 39.21 16.59 70
Figure 6. Mean key child volubility with 95% CI in families with and without older siblings
5.2.1.2 Female adult segment duration
When monthly group comparisons were conducted for neonatal health-related background variables in female adult segment durations, no statistically significant group differences were detected in any points of measure between children born with and without complications (see Appendix 4 for all statistical information). In addition, no constant group differences were found between groups differing in pretermity or birth weight, although single statistically significant differences were found. Preterm children at the age of six months heard on average 44.3% less speech from female adults than full-term children, and at the age of 10 months, LBW twins heard on average 31.4% less speech from female adults than NBW children. In addition, twins cared at bedside were found to hear more female adult speech in all but one recordings, although the difference reached statistical significance only for 10- and 12-month recordings. At 10 months, children cared at bedside heard on average 49.6% more speech from female adults, when compared with children cared at a neonatal ward. However, the 12-month point of measure had markedly unequal groups due to missing values and cannot be further discussed.
When group comparisons of neonatal background variables were conducted with the pooled data, preterm and full-term twins heard a similar amount of speech from female adults to twins born with and without birth complications (see Table 12). In this data, LBW twins heard on average 17.8% less speech from female adults than NBW twins;
this difference showed a medium practical significance. In addition, in families of twins who were cared at bedside during the neonatal treatment, twins heard on average 15.6%
more speech from female adults than twins who had been treated at a neonatal ward.
However, this difference showed small practical significance.
Table 12. Information on central decencies, dispersion, and statistical difference for group comparisons of the meaning of neonatal background variables to female adult volubility
Variable groups M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin preterm 37.51 15.77 38.43 27.56 55 2372.00 -0.86 0.93 0.01 Twin full-term 38.42 16.47 38.11 22.61 87
Twin LBW 34.20 17.52 32.85 23.07 68 1745.00 -3.15 <0.001 0.31 Twin NBW 41.62 13.96 41.87 17.11 74
No birth complications 38.67 15.40 38.11 20 .43 93
2208.00 -0.30 0.76 0.03 Born with
complications 36.92 17.60 38.50 29.64 49
Cared at bedside 40.14 15.33 39.80 19.85 95 1665.00 -2.46 0.01 0.25 Cared at a ward 33.88 17.10 33.97 24.30 47
In the monthly inspection for the difference of female adult segment durations between demographic background variables, no statistically significant differences were found for groups differing in birth order, having older siblings or the father’s education in any points of measure. A single statistically significant group difference was found between the female adult speech heard by girls and boys: on average, twin girls heard 42.9% less female adult speech than twin boys in the recordings obtained at 7 months.
In addition, the mother’s education showed a constant difference between the lower- and higher-educated mothers, although the difference reached statistical significance only in three points of measure. Children of mothers with a lower education heard female adult speech on average 40.5% more at seven, 47.5% more at nine, and 46%
more at ten months of age than children of mothers with a higher education.
Table 13. Information on central decencies, dispersion, and statistical information for group comparisons of the meaning of demographic background variables to female adult volubility
Variable groups M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin A 37.66 16.46 37.26 25.55 71 2441.00 -0.35 0.72 0.03
Twin B 38.48 15.95 38.50 22.62 71
Twin girl 36.05 17.47 36.04 23.50 86 1889.00 -2.17 0.03 0.22 Twin boy 41.17 13.45 42.47 19.47 56
No older siblings 37.72 15.90 38.10 23.28 71 2458.00 -0.26 0.80 0.02 Older siblings 38.41 16.51 38.50 24.72 71
Mom lower education 44.24 15.43 44.32 15.46 74 1300.00 -4.97 <.0001 0.48 Mom higher education 31.35 14.21 30.78 17.99 68
Dad lower education 38.69 16.03 39.57 23.07 72 2282.00 -0.16 0.87 0.09 Dad higher education 37.43 16.37 35.83 23.31 70
In the pooled data, boy twins heard on average 12.4% more female adult speech than girl twins, but the difference again yielded a small practical significance (Table 13). There was no statistically significant difference in female speech heard by twins for twin As and Bs. Twins with older siblings and first-born twins heard a similar amount of female adult speech, and differences in the level of the father’s education did not yield significant differences in the total duration of female adult speech. However, twins, who were living in families with the mother having a lower degree, heard statistically significantly more female adult speech than twins living in families with the mother with a higher education.
This difference was found to be of medium practical significance.
5.2.1.3 Male adult segment duration
No statistically significant group differences were found in male adult segment durations of pretermity, birth weight, or birth complications in any points of measure in monthly inspections (see Appendix 6 for all statistical information). The only statistically significant difference for neonatal health-related background variables was found between monthly group comparisons in male adult segment durations conducted between children cared at bedside and children cared at a ward in ten-month recordings (Appendix 6). In group comparisons, children cared at bedside heard on average 58%
more male adult speech than children cared at a neonatal ward.
The difference was also evident in the comparisons conducted using the pooled data:
children cared at bedside heard on average 20.8% more male adult speech when compared with twins cared at a neonatal ward (Table 14). However, the difference in male adult volubility showed a small practical significance between the two groups. No other statistically significant group differences were detected in neonatal background variables related to birth weight, complications at birth, or pretermity.
Table 14. Information on central decencies, dispersion, and statistical difference in group differences in neonatal health-related background variables and male adult volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin preterm 17.86 8.92 19.35 13.35 55 2128.00 -1.11 0.27 0.11 Twin full-term 18.01 13.51 13.73 15.03 87
Twin LBW 17.59 12.71 15.27 13.46 68 2345.00 -0.70 0.49 0.07 Twin NBW 18.27 11.20 14.01 15.40 74
No birth complications 18.81 12.32 16.24 13.84 93
1938.00 -1.46 0.14 0.15 Born with
complications 16.30 11.02 11.83 14.99 49
Cared at bedside 19.27 12.20 17.37 15.33 95 1732.00 -2.13 0.03 0.22 Cared at a ward 15.26 10.92 13..3585 12.82 47
In monthly inspections of group comparisons conducted with demographic variables, no statistically significant group differences were detected in the birth order or in the mother and father’s education in male adult segment durations (see Appendix 7 for statistical information). However, a constant group difference was found in mean male adult segment durations of twins with and without an older sibling, although the difference reached significance only in the measure points of 6 and 10 months. At six months, twins with older siblings heard on average 44% more male adult speech, and at ten months, on average 53% more male adult speech than twins with no older siblings.
The difference between first-born twins and twins with siblings was also evident in the pooled data. In families with older siblings, twins heard on average 28.3% more speech from male adults than in families with first-born twins (Table 15). In addition, in this data, twins living in families where the mothers had a Master’s degree or a higher education heard on average 27.0% less male adult speech when compared with families where the mothers had lower educational. However, no statistically significant group differences were found at the level of the whole data in male adult volubility for twin As and Bs, for girl and boy twins, or for groups differing in the father’s education.
Table 15. Information on central decencies, dispersion, and statistical differences in group comparisons of the effect of demographic background variables to male adult volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin A 17.48 11.52 14.40 14.22 71 2447.00 -0.30 0.75 0.03
Twin B 18.41 12.35 15.28 14.16 71
Twin girl 19.37 13.01 16.61 13.96 86 2030.00 -1.58 0.12 0.16
Twin boy 15.75 9.69 11.86 14.31 56
No older siblings 14.99 9.53 11.83 13.12 71 1764.00 -3.09 p<.001 0.30 Older siblings 20.90 13.30 19.13 14.76 71
Mom lower education 20.61 12.89 19.40 14.77 74
1817.00 -2.85 p<.001 0.28 Mom higher
education 15.04 10.05 12.89 12.80 68
Dad lower education 17.12 10.33 13.63 13.92 72 2282.00 -0.97 0.33 0.09 Dad higher education 18.79 13.36 15.94 14.64 70
5.2.1.4 Other child segment duration
When other child segment durations were inspected in monthly group comparisons of neonatal health-related background variables, no statistically significant group differences were detected in low birth weight (LBW) and normal birth weight (NBW) children or in children born with or without complications (Appendix 8). In this data, two constant group differences were detected. Full-term children and children cared at bedside heard more other child speech or speech-like vocalizations than preterm
children or children cared at a neonatal ward, although the differences reached statistical significance at six and seven months for preterm and full-term children and statistical significance at 10 months – for children cared at bedside and children cared at a ward.
Table 16. Information on central decencies, dispersion, and statistical differences in group comparisons of the effect of the neonatal health-related background variables to other child volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin preterm 22.08 9.34 21.64 17.66 55 1657.00 -3.08 0.002 0.31 Twin full-term 28.76 11.41 27.29 17.08 87
Twin LBW 27.07 12.58 25.75 16.97 68 2394.00 -0.50 0.618 0.05 Twin NBW 25.35 9.58 24.79 12.03 74
No birth
complications 27.32 11.81 25.92 15.77 93
1977.00 -1.29 0.196 0.13 Born with
complications 24.00 9.37 22.89 13.96 49
Cared at bedside 27.99 10.97 27.53 14.35 95 1550.00 -2.96 0.003 0.31 Cared at a ward 22.51 10.58 19.95 12.73 47
When group comparisons were conducted with pooled data, two neonatal health-related group comparisons reached statistically significant differences. Preterm twins heard on average 23.2% less other child speech or speech-like vocalizations than twins born full-term; this was shown to have medium practical significance (Table 16). No statistically significant group differences were found for other child volubility between LBW and NBW twins or twins born with or without birth complications.
Figure 7. Mean segment durations and 95% confidence intervals of other child segments in families with and without older siblings
When group comparisons were conducted for demographic background variables within monthly inspections, no statistically significant group differences were detected in birth order, gender, or parental education in any points of measure (Appendix 9).
However, a constant group difference with statistically significant differences in five out of seven points of measure was found for other child volubility in families with and without older siblings (see Figure 7).
In comparisons conducted with demographic background variables with the pooled data, no statistically significant group differences were detected in maternal and paternal education, nor for twins’ birth order. However, statistically significant group differences were found in other child segment durations between girl and boy twins and between twins with and without siblings. In the whole data, twin girls heard on average 20.8%
more other child speech or speech-like vocalizations than boy twins; the difference was found to be statistically significant with small to medium practical importance (Table 17). In addition, twins with older siblings heard on average 37.8% more of other child speech or speech-like vocalizations when compared with first-born twins. This difference showed a large practical significance.
Table 17. Information of central decencies, dispersion and statistical for group comparisons of the effect of demographic background variables to other child volubility
Variable group M SD Mdn IQR N U Z p 𝑟𝑟𝑏
Twin A 25.99 11.91 23.08 15.57 71
2405.00 -0.47 0.637 0.05
Twin B 26.35 10.33 25.68 14.59 71
Twin girl 28.51 11.65 27.41 15.96 86
1707.00 -2.93 0.003 0.29 Twin boy 22.59 9.23 21.96 13.28 56
No older siblings 20.07 7.97 19.05 9.46 71
830.00 6.90 <.0001 0.67 Older siblings 32.28 10.46 31.72 12.25 71
Mom lower education 25.52 10.13 25.31 13.31 74
2394.00 -0.50 0.618 0.05 Mom higher education 26.89 12.13 25.28 17.53 68
Dad lower education 26.32 9.08 26.59 12.87 72
2272.00 -1.01 0.31 0.10 Dad higher education 26.02 12.94 23.06 17.73 70