Secondary Logo

Journal Logo

Validity and Reliability of the New Handball-Specific Complex Test

Schwesig, René1; Koke, Alexander2; Fischer, David1,3; Fieseler, Georg4; Jungermann, Philipp4; Delank, Karl-Stefan1; Hermassi, Souhail5

Journal of Strength and Conditioning Research: February 2016 - Volume 30 - Issue 2 - p 476–486
doi: 10.1519/JSC.0000000000001061
Original Research

Schwesig, R, Koke, A, Fischer, D, Fieseler, G, Jungermann, P, Delank, K-S, and Hermassi, S. Validity and reliability of the new handball-specific complex test. J Strength Cond Res 30(2): 476–486, 2016—The purpose of this study was to determine the intraobserver reliability (IR) of the handball-specific complex test (HBKT) and the validity of the HBKT and nonspecific tests. Thirty experienced players (25.7 ± 3.9 years) executed the HBKT twice (time interval: 2 days). Lactate, heart rate (HR), time, throwing velocity and number of errors were measured. Afterwards, players' match performances (MPs) in 30 matches were evaluated using video analysis to compare it with the test parameters. Resting HR between first half and second half (r2 = 0.26), standing long jump (r2 = 0.18), jump and reach (r2 = 0.16), and HR before second half (r2 = 0.14) were proven to be the most valid tests or parameters. The amounts of explained variance concerning the MP of all other tests/parameters were below 10%. Overall, 41% (12/29) of the parameters showed a high relative intraclass correlation coefficient (ICC > 0.75) and absolute coefficient of variation (CV ≤ 5%) IR. Results suggest that the HBKT can be certified with an insufficient validity and a sufficient absolute (∅CV = 11.3%) and relative (∅ICC = 0.67) IR. The reasons could be insufficient tests or insufficient score of MP. The current findings suggest that the coaches and scientists should recognize a lot of effort is necessary to measure MP and to develop valid tests. Additional research should aim to connect test and MP with each other. Before a coach applies a test, he should thoroughly check whether the test is valid (gold standard: MP) and reliable. The frequent and long-term test application (very common argument of the coaches in practice) is not a proof of validity.

Departments of 1Orthopaedic and Trauma Surgery; and

2Sports Science, Martin Luther University, Halle-Wittenberg, Halle, Germany;

3Institute for Applied Training Science Department Technique-Tactics, Leipzig, Germany;

4Center for Orthopedic Surgery (ZOC), Muenden, Germany; and

5Tunisian Researches Laboratory, Sport Performance Optimization, Tunis, Tunisia

Address correspondence to Souhail Hermassi,

Back to Top | Article Outline


Team Handball (TH) is an intensive form of intermittent physical activity and a fast paced match of defensive and offensive action that includes specific movements and repeated explosive muscular contractions required for sprinting, jumping, turning, changing pace, and ball throwing (24,45). The level of performance in modern TH is determined by the players' technical, tactical, psychological/social, and physical characteristics. All these elements are of high importance in TH and also closely interlinked making TH a particularly complex type of sport (26,27).

Team Handball games are characterized by repeated periods of intense anaerobic activity (20), and it is thus logical to evaluate a player's overall ability in terms of his or her tolerance of repeated bouts of intensive exercise. However, it is unclear which part of variance is explained by those factors regarding match performance (MP). Despite the knowledge of these factors, only conditional tests (e.g., 30-m sprint test, countermovement jump, squat jump test, vertical jump test, abdominal strength, Yo-Yo Intermittent Endurance Test, and repeated sprint ability) are normally a part of the performance diagnostic (1,20,24,29).

Furthermore, there are a lot of studies concerning training in handball (4,6,12,13,18,31,39) and contributions of conditional and anthropometric characteristics of professional handball (5,14,20,24–26,30,42,44). However, handball-specific skills (e.g., dribbling, throws, passing, catching the ball, and feints) are not taken into account within a complex test design. Moreover, the different demands of the playing positions (wings, pivots, backs, goalkeepers) are also unconsidered in these tests.

Therefore, the scientific and practical situation concerning handball-specific performance diagnostic is poor and unsatisfactory. Currently, we are not able to measure the MP (first step). But there are no alternatives to this gold standard because it is the practical and key question for coaches and scientists on the way to valid tests (second step) and sufficient handball-specific training programs (third step). We can observe in science and practice a great contradiction between the often-described complex requirement profile of TH and the nonspecific and noncomplex performance diagnostic however. There are few publications within the last 16 years regarding validity or reliability of tests used in TH (7,9,20–25,36,39,41–43,45). Merely, Massuca et al. (24) and Wagner et al. (45) validated performance-based test for TH.

To date, little is known about the relationship between MP and test performance. Match analyses showed that TH involves a great deal of intermittent high-intensity activities that are undertaken by players throughout the game. To our knowledge, only 1 (45) comparable match-based performance test for TH does exist. Thus, the purpose of this study was to (a) to develop a handball-specific complex test (HBKT), (b) to validate this test, and (c) to prove the intraobserver reliability of the HBKT. Information in this context may result in great interest for training prescription and talent evaluation in TH.

Back to Top | Article Outline


Experimental Approach to the Problem

This study was designed as longitudinal and prospective investigation to examine the assessments of physical performance on the specific profile of sports. However, most of the tests in handball are unspecific and not comprehensively evaluated (i.e., lacks information about their reliability and validity). Before the development of a handball-specific test (second step), it is inalienable to parameterize the MP (first step). Therefore, we simultaneously developed an MP score (dependent variable/gold standard) and a new HBKT (independent variable) (33). The parameterization of the MP (key issue) is the precondition for development and validation of tests. The assessments of physical performance are the basis for development and evaluation of training programs (approach to the problem). The described study is part of a comprehensive study on performance analysis in handball. Other aims are to assess the effectiveness of the preseason (examination 2 vs. examination 3) using the HBKT (33) and to evaluate the strength and flexibility of the shoulder for injury prognosis (11). For this purpose, we tested 2 handball teams of the Third German League in 3 examinations during the preseason (July and August 2013): baseline, 48 hours after baseline, 6 weeks after baseline (end of the preseason).

Regarding the reliability study, the first 2 examinations were compared. The first examination was also used for the validation of the HBKT, the nonspecific handball tests, and the evaluation of the predictive value of the internal and external rotation in the shoulder. The players were closely observed during the following season (September 2013 to May 2014). Each player (wings, pivots, backs, or goalkeepers) was evaluated during match play using video recording and subsequently performing match analysis (Figure 1).

Figure 1

Figure 1

Testing of players was performed at the same time of the day (5–9 PM), and under the same experimental conditions, at least 3 days after the most recent competition. All tests took place during the preseason (first day, third day, and 6 weeks later) and were conducted on an indoor handball court, where ambient temperature ranged from 10 to 12° C. To reduce the interference of uncontrolled variables, players maintained their normal intake of food and fluids during the trial. However, they abstained from physical exercise for 1 day before testing, drank no caffeine-containing beverages in the 4 hours preceding testing, and ate no food for 2 hours before testing. Additionally, the athletes should drink 2 liters of water (only water) during the last 2 hours for the test.

Test time and investigators were identical (observational equivalence). On the day between the first and second examination, the athletes completed a 45-minute continuous run with a load limit of 75% of maximum heart rate in the HBKT. In detail, the examination proceeded on the test day as follows:

  • Presentation of the study design and the study objectives
  • Test presentation (oral, video) and test demonstration
  • 10-minute rest period in a sitting position
  • Resting heart rate and resting lactate measurement
  • 15-minute warm-up of the entire team
  • Individual test execution (time per athlete: 5–7 minutes) and simultaneous warm-up of the following individual athletes
  • 14-minute follow-up period

Thus, the duration of investigation per athlete was about 50 minutes.

Back to Top | Article Outline


In our investigation, 30 male handball professional and semiprofessional handball players were recruited (mean: 25.7 ± 3.9 years; age range: 19–33 years). All handball players were from the 2 German Third League teams (goalkeepers, n = 4; pivots, n = 6; backs, n = 13; wings, n = 7). Four handball players (13%) were left-handed. Their age and anthropometric data are listed in Table 1.

Table 1

Table 1

In line with Krüger et al. (20), the inclusion criteria for participation were the engagement of all subjects in competitive handball for at least 5 years and their participation in more than 3 training sessions per week. The players of both teams accomplished 8–10 training sessions per week during the preseason and 5–7 training sessions per week during the season. The players trained on average before the intervention period 11.9 (±3.8) h·wk−1. The training included 7.5 (±3.2) hours for handball bouts and 1.2 (±0.3) hours were for general strength training, which was performed at the end of some handball sessions. Both teams had the same amount of this strength training in which the subjects performed only sit-ups, push-ups, dips, squat jumps, etc. None of the participants reported any current or ongoing neuromuscular diseases or musculoskeletal injuries specific to the ankle, knee, or hip joints, and none of them were taking any dietary or performance supplements that might be expected to affect performance during the study.

All the players were at the competitive phase of their periodization. Before any participation, the experimental procedures, potential risks, and benefits of the project were fully explained to the participants, and they all provided written informed consent before entering the study. The study was approved by the local research ethics committee and conformed to the recommendations of the Declaration of Helsinki.

Back to Top | Article Outline


Handball-Specific Complex Test

The HBKT ((35), Figure 2) consists of 4 activity series (AS): agility parcours, defensive action, sprint (10, 20 m) and throw on goal parcours. The 4 AS were to be completed twice and containing 5 active pauses (3 times for 30 seconds, 35 seconds twice), which reflects the intermittent load character during the match. The 14-minute follow-up is based on the duration of the halftime (15 minutes) and is used for measuring the recovery ability of the handball players in this period. The following modifications were made during the tests:

  • Activity series 1: Prone with brief lifting of the arms off the floor.
  • First active pause (35 seconds): 10 impact throws with step against an inclined target surface (area: 1 × 1 m, distance: 3 m; slope: 60°) with measurement of throwing speed (minimum speed: 50 km·h−1).
  • Activity series 4: Measurement of throwing speed.
  • Activity series 4: Use of hurdles (height: 20 cm) for the implementation of the jump throws.
  • Activity series 4: Third throw (high jump throw after break through feint to throw hand) in the entire goal.
  • One second penalty per technical error or missed shot.
Figure 2

Figure 2

With the HBKT, the following parameters were measured: number of technical errors, time (e.g., 10-m sprint), throwing velocity, lactate, and heart rate. Furthermore, a lot of dimensions can be evaluated: speed with and without ball, anaerobic capacity (handball specific), speed endurance with and without ball, recuperativeness (handball specific), and handball-specific skills under pressure of time or accuracy (e.g., ball handling and throw accuracy).

Data acquisition was performed by means of following assessments:

  • Time: Time was recorded using photoelectric cells (AF Sport, Wesel, Germany), paced at the start (0 m) and at 10 and 20 m from the start. The AS 1, 2, 4 and the active pauses were measured manually with a stopwatch by 2 trained raters.
  • Throwing velocity: The maximum throwing velocity was determined by a Speed Check Radar (Stalker Solo 2; Stalker, Plano, TX, USA). The reliability of the radar system was checked by measuring rolling balls by the radar and checking them over a given distance using photoelectric cells. Intraclass correlation coefficient (ICC) and CV for the test were 0.92 and 3%, respectively (14).
  • Heart rate: Heart rate was measured using a real-time monitoring system (RS 400; Polar Electro GmbH, Büttelborn, Germany).
  • Lactate: Lactate measurements were performed using a lactate analyzer (Super GL easy; Dr. Müller Gerätebau GmbH, Freital, Germany).
  • Number of technical errors: manually (3 trained raters, majority decision).

Nonspecific handball tests were only conducted with 1 handball team (n = 14) on examination 1.

Back to Top | Article Outline

Standing Long Jump (Broad Jump)

The athlete stands behind a line marked on the ground with feet slightly apart. A 2-foot takeoff and landing is used, with swinging of the arms and bending of the knees to provide forward drive. The subject attempts to jump, as far as possible, landing on both feet without falling backwards. Best of 3 attempts is used. Parameter: distance in centimeter (distance start line/closest heel).

Back to Top | Article Outline

Jump and Reach (Vertical Jump)

The athlete stands side on to a wall and reaches up with the hand closest to the wall. Keeping the feet flat on the ground, the point of the fingertips is marked or recorded. The athlete then leaps vertically as high as possible using both arms and legs (countermovement allowed). Attempt to touch the wall at the highest point of the jump. Best of 2 attempts is used. Parameter: jump score in centimeter (distance standing reach height/jump height).

Back to Top | Article Outline

Three Hop Test (One-Leg Jump)

The athlete starts by standing behind a line with feet in step position. When ready, they are to perform 2 consecutive one-leg broad jumps non-stop using a forward and a vertical jump style. They are allowed to use their arms assist the explosive movement and for balance. The landing is parallel on both feet. Parameter: jump distance in centimeter (best of 2 attempts each leg − distance start line/closest heel).

Back to Top | Article Outline

Rope Skipping

The subject is to perform forward and backward rope skipping (stance: shoulder width) 15 seconds each. The inversion of direction is to be executed without a break. Parameter: number of hops (best of 2 attempts).

Back to Top | Article Outline

Eight-Shaped Dribbling Circuit

The player starts in step position (right hander: left foot at front; left hander: right foot at front) at the start line and runs through a 3 × 5 m circuit marked by 5 poles (8 shaped) in 3 consecutive laps. A compliant ball handling is essential. The time is electronic recorded. Parameter: time for 3 laps in seconds.

Back to Top | Article Outline

Shuttle Run Test (Beep Test)

This test involves continuous running between 2 lines 20 m apart (touch lines of the handball field) in time to recorded beeps. The subjects stand behind one of the lines facing the second line and begin running when instructed by the recording. The subject continues running between the 2 lines turning when signaled by the recorded beeps. After 1 minute (continuing), a sound indicates an increase in speed. Running speed as to be adjusted at the given pace. The test is stopped if the subject fails to reach the line (within 2 m) for 2 consecutive ends after a warning. Parameter: distance in meter.

Back to Top | Article Outline

Parameterization of the Match Performance

We calculated a match performance score (MPS) for field players and goalkeepers as followed: MP = addition of positive actions (e.g., goals, assists, and steals) and subtraction of negative actions (e.g., technical errors, missed throws, and time penalty). The calculation for goalkeepers was different (percentage caught throws based on the total number of throws). Hence, the higher the calculated score the higher the MP of the athlete. The MPS was the dependent variable within the linear regression model. Test parameters were used as independent variable.

Back to Top | Article Outline

Statistical Analyses

All statistical analyses were performed using SPSS version 22.0 for Windows (SPSS, Inc., Chicago, IL, USA). Descriptive statistics (mean, SD, 95% CI) were ascertained for the HBKT parameters. Mean differences (examination 1 vs. examination 2) of HBKT parameters were tested using a general linear model. Differences between mean values were considered statistically significant if p values were less than 0.05 and partial eta-squared (η2) values were higher than 0.10. Because of the small number of cases (e.g., position-specific analysis, Table 1), decisions of significance were made primarily based on η2 values. The statistical power was calculated using the GPOWER software (10). The statistical power for all the statistical tests was 0.95 (α error) and 0.80 (β error). The effect size for the correlation coefficient was large (0.65). Therefore, 30 subjects were necessary for a sufficient reliability analysis. Relative (ICC) and absolute (CV) reliability were calculated and interpreted based on Shrout and Fleiss (38), Hopkins (17), and Buchheit et al. (3), respectively. The ICC indicated excellent reliability if the value was above 0.75, fair-to-good reliability between 0.40 and 0.75, and poor reliability when less than 0.40 (38). The typical or random error of measurement (within-subject variation for 2 sessions) was expressed as a CV (in percentages) and derived from log-transformed data (16,17). Based on numerous earlier studies (8,15,19), a CV < 5% was set as the criterion to declare a variable as reliable. The 95% confidence intervals were calculated for each CV and ICC. The validation was performed using linear regression analysis (model: inclusion). The test parameters were used as independent variable and the MPS was the dependent variable.

Back to Top | Article Outline



From examination 1 to examination 2, improvements in almost all parameters were observed in Table 2.

Table 2

Table 2

This does not include 3 parameters in the first half (agility, defensive action, overall throwing time). Twenty-four percent of the parameters (8/33) showed significant differences (p ≤ 0.05 and η2 > 0.10) comparing the 2 sessions. The marked difference between the first half (5/10) and the second half (9/10) is remarkable.

The results of the reliability analysis were mentioned in Table 3.

Table 3

Table 3

On the basis of the cut points for the relative (ICC > 0.75) and absolute (CV ≤ 5%) reliability, it is noted that the absolute reliability (52% [17/33] reliable parameter) was greater than the relative reliability (42% [14/33] reliable parameter). Twenty-three percent (3/13) of the stress parameters and 45% (9/20) of the load parameters showed a high relative and absolute IR (ICC > 0.75 and CV ≤ 5%). Based on the confidence intervals (LL ICC > 0.75 and UL CV ≤ 5%), this proportion decreased from 36% (12/33) to 12% (4/33). For the following 4 parameters, a high reliability could be shown: heart rate after round 2, agility of first half, sprint 20 m of first half, overall throwing time of first half. The heart rate showed to be more reliable than lactate in direct comparison (∅ICC = 0.71 and ∅CV = 4.2% vs. ∅ICC = 0.65 and ∅CV = 15.1%). The average IR of the load parameters in the first half was 0.71 (ICC) and 12.2% (CV) (second half: ∅ICC = 0.60 and ∅CV = 14.3%). A low IR (ICC < 0.40 and CV > 5%) was seen in the following parameters: missed throws in the first half and missed throws in the second half.

When the parameters missed throws, the first and second half were excluded and the IR further increased (∅ICC = 0.67 vs. 0.71 and ∅CV = 11.3 vs. 10.1%). Overall, the HBKT demonstrates sufficient absolute (∅CV = 11.3%) and relative (∅ICC = 0.67) IR. Without the 2 above-mentioned parameters, the IR increases significantly (∅ICC = 0.71 and ∅CV = 10.1%). The test-retest analysis based on the Bland-Altman plots (Figures 3A, B) showed for the heart rate parameters the largest mean differences and SDs (mean = 7.7, SD = 14.0). The smallest differences between the 2 sessions were found for the load parameters in the first half (mean = −1.85, SD = 4.71).

Figure 3

Figure 3

The limits of agreement differed very significantly. They changed on average between 9.44 (load parameters of first half) and 28.0 (heart rate) and corresponded with the ICC and CV analysis (Table 3) where the parameters had the greatest reliability at the first half. After all 7 parameters (70%) in the first half had a mean of the difference whose amount was smaller than 1.

Back to Top | Article Outline


The validity of overall 40 parameters or tests was assessed through linear regression analysis (Table 4).

Table 4

Table 4

For the following 10% (4/40) parameter, an explained variance higher than 10% could be calculated:

  • Recovery heart rate (relative), end of the first half to start of the second half: R2 = 0.26
  • Standing long jump: R2 = 0.18
  • Jump and Reach: R2 = 0.16
  • Heart rate before second half: R2 = 0.14
Back to Top | Article Outline


The main finding of this study was that the IR of the HBKT must be regarded as sufficient and the validity as insufficient. This study revealed 4 highly reliable parameters (heart rate after second half, agility of first half, 20-m sprint of first half, overall throwing time of first half) and 2 parameters with a low reliability (missed throws of first and second half). Only in 2 parameters of the HBKT, the correlation of variation (recovery heart rate [relative], end of the first half to start of the second half; heart rate before the second half) was higher than 10%. Standing long jump and Jump and Reach were the most powerful unspecific tests.

Consequently, based on these results, the discussion of the external examination results is very difficult so far, because only 1 comparable test is described in the scientific literature (45). Wagner et al. (45) developed a match-based performance test for TH. Regarding test-retest reliability (interval: 7 days; n = 17), they found an ICC > 0.70 for the peak blood lactate concentration, heart rate, and ball velocity. To assess validity, Wagner et al. (45) disposed percentage running speed over the match as dependent variable (gold standard). A high correlation with the percentage running speed in the game-based performance test is shown.

Remarkably critical is the fact that the complexity of the MP is limited to the conditional parameter running speed. Souhail et al. (39) investigated the direct validity of the Yo-Yo Intermittent Recovery Test (level 1) in 18 adolescent handball players (14.3 ± 0.5 years) on the basis of MP. As said by Wagner et al. (45), they also reduced the MP to just 1 conditional parameter (total game distance). Yo-Yo IR performance (1.831 ± 373 m) was significantly related (r = 0.88, p < 0.01) to total game distance (1.921 ± 325 m). In contrast to Wagner et al. (45), which at least were able to specify the independent variable (game-based performance test), the specificity is missing for both of the variables (total test and total game distance) in the test design of Souhail et al. (39). These 2 examples are representative for a fundamental and wide-spread problem: despite the unanimous scientific references, that handball is characterized as a very complex sport (e.g., jogging, sprinting, backwards movement, sideways movement, jumps, throws, steals blocks, changes of direction, one-on-one-situations, and feints), the MP is literally reduced to conditional parameters (total game distance or relative running speed). It is not surprising that 2 conditional parameters highly correlate with each other. Thus, this is not a proof of validity, because neither Yo-Yo IR1 nor the parameterization of MP (distance, speed) meets the requirements of the sport handball. In close cooperation with Souhail et al. (39), the aim of our group is to address these issues and to seek meaningful solutions to make the performance analysis in handball more powerful.

Buchheit et al. (3) investigated the reliability of a repeated shuttle-sprint and jump ability (RSSJA) test. Semiprofessional handball players (n = 14) acted as sample group. The time interval between 2 tests was 7 days. The CVs ranged from 1.0% (sprint) and 2.9% (jump) and thus were often lower than in the HBKT (range: 1.3–68.4%). The indices based on the mean and best values, however, showed significantly higher CVs (21.2–34.8%). Regarding the previously described research, Buchheit et al. (3) discuss their findings exclusively sport unspecific. They point to similar evidence for repeated isolated sprint tests from other team sports. Sheppard et al. (37) found in male elite volleyball players a CV of 2.4% for the repetitive sprint ability. Impellizzeri et al. (19) found a CV of 0.8% for professional football players (n = 22). Sattler et al. (32) addressed volleyball-specific jumping procedures and tests. They found the highest reliability for specific jumping tests (ICC range: 0.93–0.97; CV range: 2.1–2.8%).

Mirkov et al. (28) evaluated the reliability of soccer-specific field tests. Most often, the tests revealed high ICCs (i.e., >0.80) and small interindividual variations (CV < 4%). Regarding the throwing-in and standing-kick tests, which showed a markedly lower reliability, direct measurement of the ball velocity (e.g., with standard radar gun in the HBKT) is recommended (28).

In accordance with Buchheit et al. (3), Spencer et al. (40) recommend as a result of their investigation not to use indexes as these go hand-in-hand with a reduction of reliability (CV: 14.9%, CI: 10.8–31.3%). Instead, the total sprint time (total of 6 runs) should be used. The lower reliability of HBKT compared with RSSJA is probably explained by its higher complexity. A higher test complexity implies more grades of freedom in the performance of the test.

Krüger et al. (20) showed based on parameter throw rate that the position-specific differences increase with the complexity of the test task. The mean throw velocity was between 72.7 and 90.7 km·h−1, depending on type of throw and position of the player, and was comparable with this study (range: 75.3–94.9 km·h−1).

In accordance with this study, Impellizzeri et al. (19) defined also a minimum interval between the 2 test days of 48 hours. Numerous authors (2,15,29) examined the test-retest reliability of different agility tests (hexagon test, repeated-modified agility test, agility T-test). They also used a time interval of 2 days (2,15) or 1 week (29) and reported excellent reliability for the agility tests (ICC > 0.82).

From a test-methodological perspective, it can be noted that there is currently no uniform standard regarding the time interval used in reliability studies (33,34). A small time window between test and retest promotes learning effects and thus implies a low reliability, whereas too large time windows contain development processes due to exercise, illness, and injury. This also would result in a lower reliability. In this respect, the comparability of the test results is affected not only by the lack of substantive agreement (handball specific, complex) but also by different time intervals (2–7 days).

The validation of a test is the first step in the process of evaluation. In this regard, the measure of the MP (dependent variable) is the challenge to take at present. The MP can be described as multifactorial and is roughly subdivided into offensive and defensive actions. Here, the problem is to measure the individual performance (offensive action: goals, assists, one-on-one balance; defensive actions: steals, blocks, one-on-one balance, penalties) and to put them in a sensible overall score. This purpose is hindered by the position and game system dependency of every single action. However, the test validity is highly dependent on the quality of the parameterization of the MP.

Key factors (e.g., influence of opponents, psychological distress during competition, and match ability) are also disregarded in the HBKT as they elude standardization (control). However, the HBKT lies regarding its motion-structural and temporal intensity profile significantly closer to the requirements of the sport as previously used established conditional tests (e.g., sprint tests, jump force tests, treadmill test levels, and shuttle run tests). Like any skill test, the HBKT also contains learning and adaptation effects as evidenced by the performance improvements from examination 1 to examination 2.

The observational equivalence regarding the prestart condition could not be fully ensured because the athletes were tested in succession, and the warm-up period was individually performed and planned. An individual standardized warm-up (24) is generally recommended but usually not feasible in handball due to spatial limitations.

Knowing the maximum test requirements, the retest showed a more efficient division of power on the part of the athletes. Indications are the reduced metabolic stress (parameter: lactate) in the postload period and the small deteriorations in round 1 in the AS 1, 2, and 4 (agility, defensive action, throw on goal) for examination 2 (Table 3). The AS 1, 2, and 4 were manually measured with a stopwatch by 2 trained investigators. Vicente-Rodriguez et al. (42) demonstrated that manual measurements by a trained investigator using a stopwatch are a valid method to assess assurance speed and agility fitness testing in adolescents. They pointed out that researchers must be trained to minimize measurement error.

Back to Top | Article Outline

Practical Applications

At present, there are no reliable and valid handball specific tests. The present HBKT also turned out to be inefficient, especially regarding its validity. As a result, it seems as if there is no scientific legitimating for specific performance diagnostics in handball. Test procedures are reduced to conditional parameters that are easy to assess but highly unspecific. However, it was not been possible to make the complexity of the sport (MP) measurable and reflect it in a sensible score. To resolve this problem, we propose the following procedure and order:

  • Parameterization of the MP: development of a score that reflects the MP metrically and acts as the dependent variable (gold standard) in the process of validation.
  • Handball-specific complex test: subsequently, there is to develop a highly complex and handball-specific test that is based on the evidence of the profile handball.
  • Validation: validation of the test parameters (point 2) based on MP (point 1). The MP should be measured prospectively for at least half of the season.
  • Iterative loops: an adjustment of the test procedure and a revalidation might be necessary.
  • Proof of reliability.
  • Reference data collection (position dependent and league dependent).
  • Design of handball-specific training routines.

For coaches, we suggest to thoroughly check the validity and reliability of any tests. In practice, we can often observe the argumentation of coaches “the test is already used for a long time and the other teams conduct the same tests.” But, these are not proofs of validity.

However, the described procedure is also valid and useful for similar Olympic sport ball games (e.g., soccer, hockey, football, basketball, and volleyball).

Back to Top | Article Outline


The authors would like to thank Katharina Sannemüller, Markus Becker, Andreas Wölfel and Prof. Karim Chamari for support regarding data collection and support help. The authors thank the athletes and coaches for their enthusiastic participation in the study. The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association (NSCA).

Back to Top | Article Outline


1. Andrade M, Fachina RJ, Cruz W, Benedito-Silva AA, Da Silva AC, De Lira CA. Strength field tests performance are correlated with isokinetic strength of shoulder rotator muscles in female handball players. J Sports Med Phys Fitness 54: 403–409, 2014.
2. Beekhuizen KS, Davis MD, Kolber MJ, Cheng MSS. Test-retest reliability and minimal detectable change of the hexagon agility test. J Strength Cond Res 23: 2167–2171, 2009.
3. Buchheit M, Spencer M, Ahmaidi S. Reliability, usefulness, and validity of a repeated sprint and jump ability test. Int J Sports Physiol Perform 5: 3–17, 2010.
4. Carvalho A, Mourao P, Abade E. Effects of strength training combined with specific plyometric exercises on body composition, vertical jump height and lower limb strength development in elite male handball players: A case study. J Hum Kinet 41: 125–132, 2014.
5. Chelly MS, Hermassi S, Aouadi R, Khalifa R, Van den Tillaar R, Chamari K, Shephard RJ. Match analysis of elite adolescent team handball players. J Strength Cond Res 25: 2410–2417, 2011.
6. Chelly MS, Hermassi S, Aouadi R, Shephard RJ. Effects of 8-weeks in-season plyometric training on upper and lower limp performance of elite adolescent handball players. J Strength Cond Res 28: 1401–1410, 2014.
7. Chelly MS, Hermassi S, Shephard RJ. Relationships between power and strength of the upper and lower limp muscles and throwing velocity in male handball players. J Strength Cond Res 24: 1480–1487, 2010.
8. Cormack SJ, Newton RU, McGuigan MR, Doyle TL. Reliability of measures obtained during single and repeated countermovement jumps. Int J Sports Physiol Perform 3: 131–144, 2008.
9. Debanne T, Laffaye G. Predicting the throwing velocity of the ball in handball with anthropometric variables and isotonic tests. J Sport Sci 29: 705–713, 2011.
10. Erdfelder E, Faul F, Buchner A. GPOWER: A general power analysis program. Behav Res Methods 28: 1–11, 1996.
11. Fieseler G, Jungermann P, Koke A, Irlenbusch L, Delank KS, Schwesig R. Range of motion (ROM) and isometric strength on shoulder joints of team handball sport athletes during playing season, Part II: Changes after midseason. J Shoulder Elbow Surg 23: 391–398, 2014.
12. Genevois C, Berthier P, Guidou V, Muller F, Thiebault B, Rogowski I. Effects of 6-week sing-based training of the external rotator muscles on the shoulder profile in high-chool elite female handball players. J Sport Rehabil 23: 286–295, 2014.
13. Gonzales-Rave JM, Juarez D, Rubio-Arias JA, Clemente-Suarez VJ, Martinez-Valencia MA, Abian-Vicen J. Isokinetic leg strength and power in elite handball players. J Hum Kinet 41: 227–233, 2014.
14. Granados C, Izquierdo M, Ibanez J, Ruesta M, Gorostiaga EM. Are there any differences in physical fitness and throwing velocity between national and international elite female handball players? J Strength Cond Res 27: 723–732, 2013.
15. Haj-Sassi R, Dardouri W, Gharbi Z, Chaouachi A, Mansour H, Rabhi A, Mahfoudhi ME. Reliability and validity of a new repeated agility test as a measure of anaerobic and explosive power. J Strength Cond Res 25: 472–480, 2011.
16. Hopker J, Myers S, Jobson SA, Bruce W, Passfield L. Validity and reliability of the Wattbike cycle ergometer. Int J Sports Med 31: 731–736, 2010.
17. Hopkins W. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
18. Ignjatovic AM, Markovic CM, Radovanovic DS. Effects of 12-week medicine ball training on muscle strength and power in young female handball players. J Strength Cond Res 26: 2166–2173, 2012.
19. Impellizzeri FM, Rampinini E, Castagna C, Bishop D, Ferrari Bravo D, Tibaudi A, Wisloff U. Validity of a repeated-sprint test for football. Int J Sports Med 29: 899–905, 2008.
20. Krüger K, Pilat C, Ueckert K, Frech T, Mooren FC. Physical performance profile of handball players is related to playing position and playing class. J Strength Cond Res 28: 117–125, 2014.
21. Krustrup P, Mohr M, Amstrup T, Rysgaard T, Johansen J, Steensberg A, Pedersen PK, Bangsbo J. The Yo-Yo intermittent recovery test: Physiological response, reliability, and validity. Med Sci Sports Exerc 35: 697–705, 2003.
22. Lidor R, Argov E, Daniel S. An exploratory study of perceptual-motor abilities of women: Novice and skilled players of team handball. Percept Mot Skills 86: 279–288, 1998.
23. Lidor R, Falk B, Arnon M, Cohen Y, Segal G, Lander Y. Measurement of talent in team handball: The questionable use of motor and physical tests. J Strength Cond Res 19: 318–325, 2005.
24. Massuca LM, Fragoso I, Teles J. Attributes of top elite team-handball players. J Strength Cond Res 28: 178–186, 2014.
25. Matthys SP, Fransen J, Vaeyens R, Lenoir M, Philippaerts R. Differences in biological maturation, anthropometry and physical performance between playing positions in youth team handball. J Sports Sci 31: 1344–1352, 2013.
26. Michalsik LB, Aagaard P, Madsen K. Locomotion characteristics and match-induced impairments in physical performance in male elite team handball players. Int J Sports Med 34: 590–599, 2013.
27. Michalsik LB, Madsen K, Aagaard P. Match performance and physiological capacity of female elite team handball players. Int J Sports Med 35: 590–599, 2014.
28. Mirkov D, Nedeljkovic A, Kukolj M, Ugarkovic D, Jaric S. Evaluation of the reliability of soccer-specific field tests. J Strength Cond Res 22: 1046–1050, 2008.
29. Munro AG, Herrington LC. Between-session reliability of four hop tests and the agility T-test. J Strength Cond Res 25: 1470–1477, 2011.
30. Póvoas SC, Seabra AF, Ascensão AA, Magalhães J, Soares JM, Rebelo AN. Physical and physiological demands of elite team handball. J Strength Cond Res 26: 3365–3375, 2012.
31. Rousanoglou E, Noutsos K, Bayios I, Boudolos K. Ground reaction forces and throwing performance in elite and novice players in two types of handball shot. J Hum Kinet 40: 49–55, 2014.
32. Sattler T, Sekulic D, Hadzic V, Uljevic O, Dervisevic E. Vertical jumping tests in volleyball: Reliability, validity and playing-position specifics. J Strength Cond Res 26: 1532–1538, 2012.
33. Schwesig R, Becker S, Fischer D. Intraobserver reliability of posturography in healthy subjects. Somatosens Mot Res 31: 16–22, 2014.
34. Schwesig R, Fischer D, Becker S, Lauenroth A. Intraobserver reliability of posturography in patients with vestibular neuritis. Somatosens Mot Res 31: 28–34, 2014.
35. Schwesig R, Fieseler G, Jungermann P, Noack F, Irlenusch L, Leuchte S, Fischer D. Longitudinal and specific analyses of physical performance in handball. Sportverletz Sportschaden 26: 151–158, 2012.
36. Schwesig R, Hartmann M, Leuchte S, Fischer D, Kuß O. Validity of complex soccer-specific field test and a non-specific sprint test—Assessments for test and match performance. Sportverletz Sportschaden 27: 75–84, 2013.
37. Sheppard JM, Gabbett T, Taylor KL, Dorman J, Lebedew AJ, Borgeaud R. Development of a repeated-effort test for elite Men's volleyball. Int J Sports Physiol Perform 2: 292–304, 2007.
38. Shrout PE, Fleiss JL. Intraclass correlations: Uses in assessing rater reliability. Psychol Bull 86: 420–428, 1979.
39. Souhail H, Castagna C, Mohamed HY, Younes H, Chamari K. Direct validity of the Yo-Yo intermittent recovery test in young team handball players. J Strength Cond Res 24: 465–470, 2010.
40. Spencer M, Fitzsimons M, Dawson B, Bishop D, Goodman C. Reliability of a repeated-sprint test for field-hockey. J Sci Med Sport 9: 181–184, 2006.
41. Sporis G, Vucetic V, Jovanovic M, Jukic L, Omrcen D. Reliability and factorial validity of flexibility tests for team sports. J Strength Cond Res 25: 1168–1176, 2011.
42. Vicente-Rodriguez G, Rey-Lopez JP, Ruiz JR, Jimenez-Pavon D, Bergman P, Ciarapica D, Heredia JM, Molnar D, Gutierrez A, Moreno LA, Ortega FB. Interrater reliability and time measurement validity of speed-agility field tests in adolescents. J Strength Cond Res 25: 2059–2063, 2011.
43. Vieira F, Veiga V, Carita AI, Petroski EL. Morphological and physical fitness characteristics of under-16 Portuguese male handball players with different levels of practice. J Sports Med Phys Fitness 53: 169–176, 2013.
44. Visnapuu M, Jürimäe T. Relations of anthropometric parameters with scores on basic and specific motor tasks in young handball players. Percept Mot Skills 108: 670–676, 2009.
45. Wagner H, Orwat M, Hinz M, Pfusterschmied J, Bacharach DW, Petelin von Duvillard S, Müller E. Testing game based performance in team handball. J Strength Cond Res 2014. Epub ahead of print.
46. Wilhelmsen K, Strand LI, Nordahl SHG, Eide GE, Ljunggren AE. Psychometric properties of the Vertigo symptom scale—Short form. BMC Ear Nose Throat Disord 8: 2, 2008.

test performance; match performance; heart rate; blood lactate; team sports

Copyright © 2016 by the National Strength & Conditioning Association.