TABLE 7.
Model parameters and summaries of linear and polynomial regression analyses of SLS-Berlin sentence characteristics predicting response times for the subsample of younger adults (<60 years).
|
Model parameters |
Model summary |
||||||
| Estimate | SE | t-value | p-value | adj. R2 | F-value | p-value | |
| Sentence ID | |||||||
| Linear | 65.08 | 13.5 | 4.83 | <0.001 | − | − | − |
| Quadratic | –1.70 | 0.40 | –4.25 | <0.001 | − | − | − |
| Cubic | 0.02 | 0.00 | 4.91 | <0.001 | 0.816/0.820* | 113.3 | <0.001 |
| Number of charactersa | |||||||
| Linear | –265.9 | 107.7 | –2.47 | <0.05 | − | − | − |
| Quadratic | 3.09 | 1.16 | 2.68 | <0.01 | − | − | − |
| Cubic | –0.01 | 0.00 | –2.62 | <0.05 | 0.700/0.687* | 59.2 | <0.001 |
| Number of syllablesa | |||||||
| Linear | 82.5 | 7.64 | 10.8 | <0.001 | 0.606/0.633* | 116.5 | <0.001 |
| Quadratic | − | − | − | − | − | − | − |
| Cubic | − | − | − | − | − | − | − |
| Number of words | |||||||
| Linear | –167.4 | 108.5 | –1.54 | 0.12 | − | − | − |
| Quadratic | 13.1 | 4.75 | 4.75 | <0.01 | 0.432/0.439* | 29.9 | <0.001 |
| Cubic | − | − | − | − | − | − | − |
| Number of long words (>6 letters) | |||||||
| Linear | 243.2 | 37.9 | 6.42 | <0.001 | 0.346/0.343* | 41.2 | <0.001 |
| Quadratic | − | − | − | − | − | − | − |
| Cubic | − | − | − | − | − | − | − |
| Number of nouns | |||||||
| Linear | 364.7 | 64.1 | 5.69 | <0.001 | 0.292/0.280* | 32.4 | <0.001 |
| Quadratic | − | − | − | − | − | − | − |
| Cubic | − | − | − | − | − | − | − |
| Flesch | |||||||
| Linear | –2.15 | 2.08 | –1.03 | 0.31 | 0.001/−0.002* | 1.06 | 0.31 |
| Quadratic | − | − | − | − | − | − | − |
| Cubic | − | − | − | − | − | − | − |
| Complexity ratingsb | |||||||
| Linear | 994.0 | 73.1 | 13.6 | <0.001 | 0.708/0.704* | 184.9 | <0.001 |
| Quadratic | − | − | − | − | − | − | − |
| Cubic | − | − | − | − | − | − | − |
| Point of decisionb | |||||||
| Linear | –134.3 | 123.0 | –1.09 | 0.28 | − | − | − |
| Quadratic | 12.5 | 5.77 | 2.17 | <0.05 | 0.415/0.401* | 28.0 | <0.001 |
| Cubic | − | − | − | − | − | − | − |
All models were checked for influential data points and outliers. aBest model fit after excluding influential cases (i.e., sentence 1) due to a Cook’s distance > 1 and high leverage. bIn a prestudy 19 participants rated the complexity (on a five-point scale ranging from one to five) and marked the point of semantic decision for all 77 sentences. *Adjusted R2 after cross-validation.