Skip to main content

Table 14 Robustness tests on unique variation explained by task measures: survey vs expert data

From: On the measurement of tasks: does expert data get it right?

Model: individual-level & expert-based task data

(1)

(2)

(3)

(4)

(5)

(6)

(7)

Abstract (Ind.)

5.7%

5.9%

7.4%

6.2%

4.4%

5.8%

4.1%

Routine (Ind.)

1.8%

1.6%

2.2%

1.9%

0.2%

3.6%

1.9%

Abstract (Exp.)

5.6%

4.4%

2.9%

4.4%

3.2%

3.7%

5.7%

Routine (Exp.)

2.3%

1.6%

1.6%

1.5%

1.8%

1.0%

1.3%

Total (Occ.)

7.9%

6.0%

4.5%

6.0%

4.9%

4.7%

7.1%

Total (Ind.)

7.5%

7.4%

9.6%

8.0%

4.6%

9.4%

6.0%

Wage/hr \(\ge 5\) & Empl. Hours \(\ge 15\)

\(\checkmark\)

      

Observations per Occupation \(\ge 100\)

 

\(\checkmark\)

     

Occupational Classification: 2-digit

  

\(\checkmark\)

    

Activities performed “often” OR “sometimes”

   

\(\checkmark\)

   

Task construction: exclude competencies

    

\(\checkmark\)

  

Task normalization a la Alda (2013)

     

\(\checkmark\)

 

2012 Sample only

      

\(\checkmark\)

Observations

26,641

25,468

27,777

28,026

27,069

27,777

13,756

  1. The displayed values represent the unique variation in log wages associated with the task measure of interest, expressed relative to the R-squared of the full model. The output provides robustness checks on the baseline model comprising individual-level and expert-based task measures, see Table 8 for reference. I use BIBB/BAuA data that has been collected in 2011–12 and 2017–18, and data from BERUFENET, covering the years 2011–13. For the BIBB/BAuA data, see Hall et al. (2020b) and (Hall et al. (2020a), respectively. For the BERUFENET data, see Dengler et al. (2014)