Dissecting\critiquing an abstract

1. Dissecting\critiquing an abstract More feedback on critiquing and validity

2. The abstract • From: • Twist, C., Gleeson, N., & Eston, R. (2008). The effects of plyometric exercise on unilateral balance performance. Journal of Sports Sciences, 26(10), 1073-1080.

3. The abstract • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance. Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training.

4. Step 1 • Read through the abstract and get a general idea of what it is about.

5. Step 2 • Identify the relationship of interest in the study, together with all dependent and independent variables

6. Step 2: what’s the relationship? • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance.Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training. Seems pretty clear this is a causal study, or at least one that is trying to ascertain cause (“effects”)

7. Step 2: what’s the relationship? • The purpose of this study was to determine the effects of plyometric exercise on unilateral balanceperformance.Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training. Also, in this case the independent… …and dependent variables are quite clear

8. Step 2: what’s the relationship? • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance. • So the relationship is expecting that P.E. will change U.B.P. in some way (note this opening sentence doesn’t say how) • It’s likely that you’d want to get an idea behind the “how” and “why” of this relationship from the lit review of the full paper in a full length paper critique

9. Step 3: external validity/sampling • First you need to get an idea of the intended scope of generalization • Look for clues in the title: • The effects of plyometric exercise on unilateral balance performance • None here – seems all populations, times and places are fair game • Look for clues in the abstract:

10. Step 3: external validity/sampling • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance. Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training. They chose, and mentioned, healthy adults. So perhaps this is one way they are limiting generalization The relationship to be considered is that plyometric training decreases balance (note that it recovers after 24hr) They are clearly interested in other settings, and other times, so these are fair game for us

11. Step 3: external validity/sampling • First you need to get an idea of the intended scope of generalization • Look for clues in the abstract: • It seems the assumption is that these findings would be equally generalizable to all healthy adults • (no limits on their sample) • They seem to want to generalize the relationship to other activities involving balance, not necessarily just unilateral (“skill-based activities”) • They seem to want to generalize to multiple or repeated bouts of plyometric training (“increased injury risk following high-intensity plyometric training”)

12. Step 3: external validity/sampling • Next you need to consider whether these are reasonable assertions • It seems the assumption is that these findings would be equally generalizable to all healthy adults • Consider the sampling technique • None apparent – so now what • Can you think of anything that would limit generality to a particular subsection of the population? • Trained/untrained? Age? Gender? A particular type of sport? • All should be considered in terms of impact on relationship of interest.

13. Step 3: external validity/sampling • Next you need to consider whether these are reasonable assertions • They seem to want to generalize the relationship to other activities involving balance, not necessarily just unilateral (“skill-based activities”) • Think of things that might affect the relationship, again • Perhaps the effect in this study emerged because the activity was unfamiliar to the subjects – perhaps a skilled activity would be more resistant to the effects • Perhaps the skip from unilateral balance to other skill based activity is not so simple (just because it affects unilateral balance, doesn’t mean it’ll affect bilateral balance)

14. Step 3: external validity/sampling • Next you need to consider whether these are reasonable assertions • They seem to want to generalize to multiple or repeated bouts of plyometric training (“increased injury risk following high-intensity plyometric training”) • Think of things that might affect the relationship, again • Perhaps the effect in this study emerged because only one bout was considered (with one bout of plyometrics, muscle soreness is quite common, with several bouts, less so – see internal validity, later) • Note that the impairment was only short term (24hrs, then “recovered”) • Perhaps result not relevant for in-season, when specific training doesn’t occur within 24hrs of competition

15. Step 4: construct validity • First, identify the constructs (what are the authors trying to measure), and the relationship of interest • Then identify the operationalizations, and whether these measure “the construct, the whole construct, and nothing but the construct” • Then consider how likely it is that departures from perfection will alter the outcome of the study

16. Step 4: construct validity Cause construct Effect construct • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance. Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training. Measures – note several do not measure UBP

17. Step 4: construct validity • First, identify the constructs (what are the authors trying to measure), and the relationship of interest • Cause: plyometric exercise • Would like to see paper to ask: is it any kind of plyometric training that is of interest, or training of a particular intensity, or over a particular duration… • Effect: unilateral balance • Would like to see paper to ask whether overall balance, or indeed overall skilled performance, is the true intended construct

18. Step 4: construct validity • Then identify the operationalizations, and whether these measure “the construct, the whole construct, and nothing but the construct” • Cause: Plyometric exercise consisted of 200 counter-movement jumps • Duration? Intensity? Mode? Variations in program? A real training bout? Is this actually plyometric training? Or just movements that simulate what the supposed effects of such training are?

19. Step 4: construct validity • Then identify the operationalizations, and whether these measure “the construct, the whole construct, and nothing but the construct” • Effect: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance • Provided this is one-legged, it seems to have face validity • How scored? Objective? • Do people perform consistently on this (reliability)? • Does it agree with other measures of stability?

20. Step 4: construct validity • Then consider how likely it is that departures from perfection will alter the outcome of the study • Cause • Do the activities constitute plyometric training? • If they don’t, then the outcome of the study is potentially irrelevant • Assuming they do constitute P.T., to what extent do plyometric programs vary? Could these variations mean that the activities performed in this paper are in effect just one type of PT, and we need to examine more varieties? • Perhaps there’s something about the number of repetitions, or the amount of concentric contraction… • Can such variations mean this one set of activities are more likely to generate changes in balance than others?

21. Step 4: construct validity • Then consider how likely it is that departures from perfection will alter the outcome of the study • Effect • Is it possible that performance relies on something other than balance? • Strength? • If it does, it could be that this strength training is just affecting strength, not balance, and so that conclusions need to be reconsidered.

22. Step 5: internal validity • First ask yourself what the design of the study is • Then ask what group of internal validity threats is typically to be considered for this particular design (consider temporal precedence, covariation of cause and effect as well if necessary) • Then ask yourself whether the alternatives are sufficiently controlled/discussed, or are in fact as likely to explain the results as the proposed cause

23. Step 5: internal validity • The purpose of this study was to determine the effects of plyometric exercise on unilateral balance performance. Nine healthy adults performed baseline measurements on the dominant limb that consisted of: a 20-s unilateral stability test on a tilt balance board, where a higher stability index represented deterioration in balance performance; isokinetic plantar flexion torque at 0.52 and 3.14 rad · s-1; muscle soreness in the calf region; and resting plantar flexion angle. Plyometric exercise consisted of 200 counter-movement jumps designed to elicit symptoms of muscle damage, after which baseline measurements were repeated at 30 min, 24, 48, and 72 h. Perceived muscle soreness of the calf region increased significantly following the plyometric exercise protocol (F4,32 = 17.24, P < 0.01). Peak torque was significantly reduced after the plyometric exercise protocol (F4,32 = 7.49, P < 0.05), with greater loss of force at the lower angular velocity (F4,32 = 3.46, P < 0.05), while resting plantar flexion angle was not significantly altered compared with baseline values (P > 0.05). The stability index was significantly increased (F4,32 = 3.10, P < 0.05) above baseline (mean 2.3, s = 0.3) at 24 h (3.3, s = 0.4), after which values recovered. These results indicate that there is a latent impairment of balance performance following a bout of plyometric exercise, which has implications for both the use of skill-based activities and for increased injury risk following high-intensity plyometric training.

24. Step 5: internal validity • First ask yourself what the design of the study is • One group, with a pre-test, a treatment, and 4 post-tests (30min, 24hr, 48hr, 72hr) • So a single group pre-test post-test non-experimental design • So what does this give us?

25. Step 5: internal validity • Then ask what group of internal validity threats is typically to be considered for this particular design Clearly, we are here – in single group threat land. This is a weak design. Don’t only think of these threats though, but also of confounds, and the possible effect of experimenter bias

26. Step 5: internal validity Consider the actual pattern of results obtained here • Then ask yourself whether the alternatives are sufficiently controlled/discussed, or are in fact as likely to explain the results as the proposed cause • The lack of a control group is troubling here – perhaps any strength training program inhibits performance on this task (construct/external issue, perhaps) • Note that the program also affected muscle soreness • Perhaps it’s note the plyometrics per se, but muscle soreness that’s the culprit (no surprise to the authors, this one) – but perhaps plyometrics do not have to cause muscle soreness?