Table 3 Responses to statements included in the Definition of Big Data domain

From: A Delphi study to build consensus on the definition and use of big data in obesity research

Big Data….

Round 1 (n = 36)

Round 2 (n = 29)

Round 3 (n = 26)

 

Agree %

Disagree %

Agree %

Disagree %

Agree %

Disagree %

1. Always has a large sample size

77.8%

22.2%

75.9%

24.1%

88.5%

11.5%

2. Always requires additional computing power

68.6%

31.4%

69.0%

31.0%

80.8%

19.2%

3. Is never collected for research purposes (i.e. there is no a priori research question)

25.7%

74.3%

21.4%

78.6%

15.4%

84.6%

4. Is always observational

40.0%

60.0%

22.2%

77.8%

23.1%

76.9%

5. Does not require specialist mathematical or data science analytical skills

12.1%

87.9%

7.1%

92.9%

8.0%

92.0%

6. Does not require specialist knowledge of database management

15.6%

84.4%

14.8%

85.2%

12.5%

87.5%

7. Does not require knowledge of computer programming

42.4%

57.6%

37.0%

63.0%

25.0%

75.0%

8. Is always digital

61.8%

38.2%

66.7%

33.3%

72.0%

28.0%

9. Does not include qualitative data

35.3%

64.7%

17.9%

82.1%

23.1%

76.9%

10. Includes government data sets

94.3%

5.7%

96.6%

3.4%

96.2%

3.8%

11. Includes cohort data sets

86.1%

13.9%

92.9%

7.1%

96.2%

3.8%

12. Includes commercial data sets

97.2%

2.8%

96.6%

3.4%

96.2%

3.8%

13. Includes routine data sets

94.4%

5.6%

96.6%

3.4%

100.0%

0.0%

14. Always includes more than one data set

16.7%

83.3%

17.2%

82.8%

15.4%

84.6%

15. Big data always has at least one of:

 • large volume (e.g. in terms of sample size, number of variables or measurement occasions),

• variety (e.g. in terms of the types of variable), or

• velocity (e.g. is generated at speed)

93.1%

6.9%

92.3%

7.7%

  1. Note: Bold % denotes that 70% consensus was achieved