Agricultural Statistics Oneliners

1. Mean, Median and Mode are Measures of Central tendency.

2. A figure obtained by dividing the sum of all variables by their total number of variables is called Averages/Arithmetic Mean.

3. Sum of deviation of items from the Arithmetic Mean (A.M.) is 0.

4. The mean affected by change in origin and scale both is Arithmetic Mean (AM).

5. Middle most value of the series is called Median.

6. Median represents the 50th Percentile.

7. The most frequently occurred item is called Mode.

8. The relationship between AM, median and Mode in asymmetrical distribution is Mode = 3 Median – 2 Mean.

9. The best measure of central tendency is Arithmetic Mean (AM).

10. The ratio of number of observations to the sum of the reciprocal of the value of the different observations is called Harmonic Mean.

11. The order of three averages for a given data is AM > GM > HM.

12. Mean applied when dealing with rate, price, and speed of a vehicle is Harmonic Mean (HM).

13. Mean applied when dealing with relative changes, e.g., bacterial growth, cell division, population, is Geometric Mean (GM).

14. The average of the sum of squares of the deviation about mean is called Variance.

15. The degree of scatterness or variation of the variable about a central tendency is Dispersion.

16. MD, SD and Variance are Measures of Dispersion/Spread.

17. Half of the interquartile range is called Quartile deviation.

18. The best measure of dispersion is Standard Deviation (SD).

19. SD is always calculated by Arithmetic Mean (AM).

20. SD ranges from 0 to ∞.

21. The difference between highest and lowest value of the series is called Range.

22. Unit less figure based on two values is Range.

23. Coefficient of variation is calculated by CV = (SD/Mean) × 100.

24. The variation used to compare the variability between two series is Coefficient of Variation (CV).

25. Which is not a measure of dispersion? Coefficient of Variation (CV) (Note: Possibly context-specific or an error).

26. The measure of the direction and degree of asymmetry is Skewness.

27. The formula of Karl Pearson’s coefficient of skewness is CSK = (Mean - Mode) / SD (formula incomplete in text).

28. Coefficient of skewness for normal distribution is 0.

29. An idea about the flatness or peakedness of the curve is called Kurtosis.

30. The term ‘Kurtosis’ was introduced by Karl Pearson (1906).

31. A curve with β2 > 3 or Y2 > 0 is called Leptokurtic curve.

32. A curve with β2 = 3 or Y2 = 0 is called Mesokurtic curve.

33. The study of association or degree and deviation between two or more variables is called Correlation.

34. Correlation lies between -1 to +1.

35. Used to measure the average relationship between two or more variables is Regression.

36. Regression coefficient is independent of Origin.

37. The distribution in which Mean > Variance is Binomial distribution.

38. The distribution in which Mean = Variance is Poisson distribution.

39. The degree of freedom of Normal distribution is n-3.

40. The term used to denote chance of happening or not happening of an event is Probability.

41. Probability is formulated by Probability = (Number of favorable events) / (Total number of events) (formula incomplete in text).

42. Probability ranges from 0 to 1.

43. The test used for comparing two means when sample size is small (up to 30) is ‘T’ test.

44. Student's t test is used when sample size is small and SD is unknown.

45. Student's t test was proposed by W.S. Gosset.

46. To test the proportions and variance, we use ‘F’ test.

47. To test the goodness of fit or homogeneity, we use CHI2 test.

48. CHI2 test was given by Karl Pearson.

49. When the calculated F is greater than table F value at 5%, the differences in treatments are considered Significant.

50. With increasing number of error degrees of freedom, table F value follows a gradually decreased trend.

51. Logical constructions of experiments where the degree of uncertainty with which the inference (result/confusion) may be defined is called Design of Experiments.

52. The objects of comparison, which an experiment tries in the field for assessing their value, are called Treatment.

53. The 3 basic principles of field experimentation are Replication, Randomization and Local control.

54. Repeated application of treatments is called Replication.

55. Allocation of treatments to different experimental units by a random process is called Randomization.

56. The principle of experimentation that eliminates human biases is Randomization.

57. Local control helps in reducing Experimental error.

58. The transformation required when data does not follow normal distribution is called Data transformation.

59. The most appropriate transformation for percentage data is Angular transformation.

60. The transformation applied when mean equals variance is Square root transformation.

61. The hypothesis under test is called Null hypothesis.

62. The variation due to uncontrolled factors is called Experimental error.

63. The error in which the hypothesis is true but our test rejects it is called Type I error.

64. Out of the two types of error in testing, the more severe error is Type II error.

65. The simplest experimental design is Completely Randomized Design (CRD).

66. The experimental design which provides maximum degree of freedom for error is CRD.

67. The design applied when experimental materials are limited and homogenous is CRD.

68. The error degree of freedom in CRD is formulated as N - t (N = total observations, t = treatments).

69. The most commonly used design is Randomized Block Design (RBD).

70. RBD is also called One way elimination of heterogeneity design / Two way classification of ANOVA.

71. When fertility gradient is in one direction, the statistical design to be used is RBD.

72. The maximum number of treatments adopted in RBD is 20.

73. In RBD, the number of blocks is equal to number of replications (b = r).

74. The error degree of freedom of RBD is formulated as (t-1)(r-1).

75. The design in which fertility gradient is in two way direction is Latin Square Design (LSD).

76. LSD is also known as Two way elimination of heterogeneity design / Three way classification of ANOVA.

77. In LSD, the number of rows, columns or treatments is equal to number of replications (r = c = t).

78. The optimum number of treatments studied in Latin Square Design is 5 to 12.

79. The error degree of freedom of LSD is formulated as (t-2)(t-1).

80. The design which provides main effects and interactions is Factorial RBD.

81. The treatment degrees of freedom for 3 factors each at 2 levels is 2³ - 1 = 6 - 1 = 5.

82. The technique of reducing the size of replication over a number of blocks at the cost of losing some information on same effect is Confounding Design.

83. Interactions are unimportant in Confounding Design.

84. Confounding Design is adopted when the number of treatments is 10.

85. If an interaction effect is confounded with all the replicates of the treatment, it is called Complete/total confounding.

86. The most appropriate design, when all factors are not of equally important in experimentation, is Split Plot Design (SPD).

87. To study two factors with different levels of precision, the design used is Split plot design.

88. The factor requiring larger units to be applied and may produce larger differences is called Main plot.

89. The error degree of freedom of SPD is formulated as D(r-1)(d-1).

90. In a split plot design with 5 levels of main plot and 4 levels of subplot treatments studied with 3 replications, the degrees of freedom for error b source is 30.

91. If sub treatments are laid out in strips, the design is called Strip Plot Design.

92. The number of error variances applied in Strip Plot Design is 3.

93. In Strip Plot Design, the component tested with higher precision is Interaction.