Text editing example (good)

Perception of energy drink effects

Bliss & Depperschmidt carried out research into the perceived physical, psychological and practical effects of energy drinks among student pilots in 20101. This article provides a meta-analysis done on the original results.

Illustration 1 shows that, overall, the students agreed that energy drinks had some negative practical effects on flying but disagreed that they posed physical or psychological effects (when attending to the standard deviation for all three probes, the majority, 68%, of students held moderate opinions, and tended to either agree or disagree, rather than strongly agree or disagree, that energy drinks brought about negative effects).

Illustration 1: Probes by frequency of responses, mean, and standard deviation
Frequency of responses Statistics
Effects Strongly disagree (1) Disagree (2) Agree (3) Strongly agree (4) Mean StdDev
Psychological 12 47 28 3 2.24 0.72
Physical 5 41 40 4 2.48 0.67
Practical 2 11 42 5 2.83 0.62
(*Statistics based on the distribution of values (frequencies x anchor's value)

When attending to the effect sizes of the differences between probes (illustration 2), there was a small delta between perception of psychological and physical effects (d = 0.36), and a medium delta between physical and practical effects (d = 0.52). However, we can advance a similar argument than in Robinson et al (20123), that such effect sizes may not reflect a similarly important difference in scores between answers. The argument goes as follows: if we take each anchor as the centroid for a particular opinion (eg, '3, Agree'), it is reasonable to assume that such estimate actually ranges midway between the immediately lower and upper anchors (eg, any group score between 2.5 and 3.4 would be considered 'Agreement'). Thus, a difference of 0.5 points on this scale (ie, a difference that can differentiate between a centroid and the next range) may be a reasonably minimum degree of expected difference before considering such difference as something noteworthy5. Following this logic, then, most of the mean differences in agreement between probes were below such threshold and, thus, rather unimportant.

Illustration 2: Probes by effect size
Comparison Means difference Effect size (d) Interpretation
Psychological vs physical 0.24 0.36
Physical vs practical 0.35 0.52
Practical vs psychological 0.59 0.88 small
[Effect size (d) based on the average standard deviation (= 0.67)]

Study's scope

The sample used comprised only student pilots from a particular location in the USA. If at all, findings may only be generalized to similar populations. This being said, the study does give direction for future research on the effects of energy drink consumption because it identifies what areas student pilots believe are affected by such consumption.

Methods

Research approach

Exploratory research into the effects of energy drinks as perceived by student pilots.

Sample

A convenient sample of 30 collegiate flight students from Oklahoma State University. The average (median) student was male, in his third year at university, had between 100 and 149 hours of flight experience, consumed between 1 and 3 energy drinks per week, experienced negative side effects (jolts, headaches, palpitations, etc) when consuming more than one energy drink a day, and had consumed an energy drink the same day he piloted an aircraft.

Variables

Bliss & Depperschmidt (2010)1 used ten Likert-type items with four anchors each (from 'strongly agree' to 'strongly disagree') to gauge the students' perceptions of the effect of energy drinks. In this article, those items have been grouped into the following probes:

• Physical effects probe (effects on the way the body operates):
• Jolt and crash (high/low energy) episodes are typical after consumption of energy drinks
• Headaches are common after consuming energy drinks
• Heart palpitations (pounding or racing) are common after consuming energy drinks
• Psychological effects probe (effects on the way the mind operates):
• Energy drinks have an effect on short term memory
• Chronic use of energy drinks can lead to other use of stimulants
• The consumption of energy drinks can be associated with risky or behavior problems
• Practical effects probe (effects on the ability to pilot an aircraft):
• Energy drinks have a (negative) effect on collegiate flight students’ ability to pilot an aircraft
• Collegiate flight students may consume an energy drink the same day they operate an aircraft

Two of the original items were omitted for analysis in this article: "Consumption of energy drinks is considered similar to consumption of coffee" was deemed trivial. And "Energy drinks are an effective and safe method to increase a collegiate flight student's mental and physical performance" was deemed too ambiguous, as it is unclear whether an answer is in regards to 'being effective', 'being safe', or both, and whether it is in regards to 'increasing mental performance', 'increasing physical performance', or both.

The item "Energy drinks have an effect on collegiate flight students’ ability to pilot an aircraft" is also ambiguous, as it is unclear whether it refers to a positive or a negative effect. However, there is an overall negative bias in the remaining items, so this one was kept here as being probably perceived as meaning "having a negative effect".

Negative items in the original study have been reversed in this meta-analysis, and the numerical ordinal values allocated to the original anchors have also been reversed as follows: '1 = SD or Strongly disagree', '2 = D or Disagree', '3 = A or Agree', and '4 = SA or Strongly agree'.

Data analysis

The original research article only provided frequencies and percentages. This meta-analysis, however, provides further descriptive statistics, namely central tendency, dispersion and effect size.

References
1. BLISS Timm J & Chad L DEPPERSCHMIDT (2010). Energy drink consumption and its effects on student pilots: perceptions of collegiate flight students4. Collegiate Aviation Review (ISSN 1523-5955), 2011, volume 29, number 2, pages 1-12.
2. DEPPERSCHMIDT Chad L, Timm J BLISS & Conrad L WOOLSEY (2010). The effect of energy drink consumption on collegiate flight students' pilot skills in a simulated flight environment. Collegiate Aviation Review (ISSN 1523-5955), 2010, volume 28, number 2, pages 1-15.
3. ROBINSON Melissa & Jose D PEREZGONZALEZ [eds] (2012). Judgements of likelihood under hypoxic conditions (descriptive statistics). Journal of Knowledge Advancement & Integration (ISSN 1177-4576), 2012, pages 271-274.
+++ Notes +++
4. This article is based on a research done in 2010 (same sample size, sample demographics and IRB application number) and partly published as Depperschmidt et al (2010)2. This note is necessary because the authors seem to be trying to pass it out as an independent research done a year later (eg, in page 2).
5. Put in perspective, 0.5 points represents a 16% change in value on a 4-anchor Likert item.

 Other interesting sites
page revision: 0, last edited: 15 Jul 2013 23:56