Kinesiologia Slovenica, 24, 1, 43-57 (2018), ISSN 1318-2269
Original article 43
Johanna Weber1	PERFORMANCE DEMANDS ON
Linda van Maanen-Coppens2 FEMALE TEAM HANDBALL Manfred Wegner3	GOALKEEPERS
ZAHTEVE GLEDE USPEŠNOSTI VRATARK V EKIPNEM ROKOMETU
ABSTRACT
Success in competitive sports is limited by constitutional, conditional, coordinative, technical, psychological and tactical factors. These factors differ by position in team handball, and the differences have yet to be sufficiently specified. Especially for female goalkeepers, the demands remain unclear. Testing for goalkeeper performance demands in handball reveals that previous values do not match those of field players. Still, the goalkeepers tested played in top leagues and apparently met game standards for their position. Consequently, the core demands of the goalkeeper position must be clarified, as well as how goalkeepers differ from field players. To this end, 654 female handball players from German leagues of all performance levels were tested for the above-mentioned factors. Each player's handball-specific expertise was used as the independent variable while the performance differences between positions were seen as dependent variables, as was goalkeeper performance. There were significant differences between goalkeeper and field player performance. Core position demands for goalkeepers were experience ("anticipation expertise"), ambidexterity, low action/state orientation after malperformance and the 30m sprint (fastest of two attempts). Results also suggest that goalkeepers should orient their performance on that of field players for the following factors; sit-ups, reaction test, jump & reach, 10 and 20m sprint, Cooper test, throwing speed, wall passing, slalom dribbling, tactics test, hope for success, net hope, self-optimization, self-impediment, lack of activation, and action/state orientation after malperformance. Coaches should implement specialized training for goalkeepers.
Key Words: performance, psychological factors, positions, goalkeepers, specialization
'Institute for Sports Science, Christian-Albrechts-University of Kiel
2Netherlandian Handball Federation (NHV), Oosterbeek 3Institute for Sports Science, Christian-Albrechts-University of Kiel
Corresponding author: Dr. Johanna Weber CAU of Kiel (Home office) Schlofistr. 14
IZVLEČEK
V tekmovalnih športih je uspeh omejen z dejavniki, povezanimi s telesno zgradbo in pripravljenostjo ter s tehničnimi, psihološkimi in taktičnimi dejavniki.
Ti dejavniki se razlikujejo glede na položaj v ekipnem rokometu, razlike pa je treba še ustrezno opredeliti. Zlasti v primeru vratark so zahteve še vedno nejasne. Testiranje zahtev glede uspešnosti vratark v rokometu je pokazalo, da vrednosti niso enake kot pri ostalih igralkah. Ne glede na to pa so vse testirane vratarke igrale v najvišjih ligah ter so očitno izpolnjevale merila za svoj igralni položaj. Zato je treba glavne zahteve za igralni položaj vratarke natančno opredeliti ter obenem določiti, po čem se vratarke razlikujejo od drugih igralk. Da bi lahko preučili zgoraj navedene dejavnike, smo testirali 654 rokometašic, ki igrajo v nemških ligah na vseh ravneh uspešnosti. Veščine vsake igralke, značilne za rokometno igro, so bile uporabljene kot neodvisne spremenljivke, medtem ko so bile razlike v uspešnosti med različnimi položaji igralk obravnavane kot odvisne spremenljivke, npr. uspešnost vratarke. Med uspešnostjo vratarke in uspešnostjo drugih igralk so bile značilne razlike. Glavne zahteve za položaj vratarke so bile izkušnje (,sposobnost predvidevanja'), obojeročnost, tendenca k nižjemu aktivnemu stanju po neuspehu ter 30-metrski šprint (hitrejši od dveh poskusov).
Rezultati kažejo tudi, da bi morale vratarke svojo uspešnost približati uspešnosti ostalih igralk v naslednjih dejavnikih: trebušnjaki, test odziva, skok in doseg, 10- in 20-metrski šprint, Cooperjev test, hitrost meta, prebijanje zidu, slalomsko preigravanje, test taktike, upanje na uspeh, splošno upanje, lastna optimizacija, oviranje samih sebe, pomanjkanje aktivnosti in tendenca k nižjemu aktivnemu stanju po neuspehu. Trenerji bi morali izvajati specializirane treninge za vratarke.
Ključne besede: uspešnost, psihološki dejavniki, položaji, vratarke, specializacija
38165 Essenrode
j.weber_@hotmail.de
+4901704100010
44 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
INTRODUCTION
In team handball there are positional differences between the players in regards to several factors (Cavala, Trninic, Jasic & Tomljanovic, 2013). Therefore players should be trained and selected to fit the specific demands of their position. It is still unclear to which extent and when specialization, position-specific selection or training should be applied.
Previous studies conducted with female players found differences between goalkeepers and field players, with goalkeepers not matching field players' values despite competing at a high level (Ignat'eva, Petracheva & Savinkov, 2002; Ignat'eva & Minabutdinov, 2014; Rogulj, Srhoj, Nazor, Srhoj & Cavala, 2005; Zapartidis, Toganidis, Vareltzis & Christodoulidis, 2009 a). It has to be mentioned that insufficient goalkeeper preparation during training may have influenced goalkeeper performance in past studies (De Castro, Sequeira & Cruz, 2011; Ignat'eva, Petracheva & Savinkov, 2002; Ignat'eva & Minabutdinov, 2014; Zapartidis, Kororos, Christodoulidis, Skoufas & Bayios, 2011). However, another reason for the difference in values between goalkeepers and field players could be that the crucial performance factors for female goalkeepers were not tested in those studies (Kajtna et al., 2011; Pori, Sibila, Justin, Kajtna & Pori, 2012). Past studies might accidentally have researched the demands of field players (Kajtna et al., 2011), probably assuming they could also be applied to goalkeepers.
Most studies so far mainly focus on male players (Karcher & Buchheit, 2014) or do not distinguish between male and female players (Wagner, Finkenzeller, Wurth & von Duvillard, 2014). However, differences between male and female players are to be expected (Marczinka, 2011) and some factors crucial to goalkeeper performance in female players have been defined. At this point, experience in play and the resulting anticipation-expertise are considered important (Schorer, 2007). So are conditional factors such as being able to quickly develop the maximum amount of force when jumping (as assumed by Pori, Justin, Kajtna & Pori, 2011). Regarding constitution, goalkeepers in the European Championships are tall and heavy with a relatively high body fat percentage (Urban, Kandrac & Taborsky, 2011). Tactics and positional play are also important (De Castro et al., 2011). In terms of psychological performance, Kajtna et al. (2011) state that male goalkeepers function in an action-orientated manner when coping with failures. Successful goalkeepers did not think about failure as long as less successful goalkeepers and concentration, fear and aggression were not particularly developed in top goalkeepers (Kajtna et al., 2011). Karcher & Buchheit (2014) analyzed literature published up to 2014 and extracted demands for male players on the different positions. It is unclear whether these are applicable to female players (Marczinka, 2011). However, some findings for female goalkeepers have been mentioned in literature (see supplemental material). Nevertheless, only few studies (Speicher, Klein Kleinoder, Schack & Mester, 2006 for cognitive speed of action; Weber & Wegner, 2016 a for constitutional factors; Weber & Wegner, 2016 b for psychological factors) have tested for connections between test results and players' actual success.
Consequently, there are only a few performance relevant factors already identified and it is not clear which other factors may be relevant in goal in female team handball. Most studies do not assess the connection between performance in testing (e. g. motor abilities) and match success (Pori et al., 2012). Most studies so far are descriptive in nature (Manchado, Tortosa, Vila, Ferragut & Platen, 2013).
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 45
As psychological demands are seen as the main performance limiting factor for goalkeepers (Kajtna et al., 2011) and are a good talent predicator (Gon^alves, Rama & Figueiredo, 2012), the exact make-up of a goalkeeper's psychological profile should also be researched. Currently, many test results for goalkeepers are caused by what is called "negative selection" when selecting for that position (Zapartidis et al., 2009 a). In youth training, players are sometimes selected for the goalkeeper position because they do not display evident handball skills or they are new to the team (Matthys, 2012; Zapartidis et al., 2009 a). Children are therefore chosen as the goalkeeper because they are overweight, tall or have motor skill deficits, while only rarely is it because a child volunteered (Sibila, Pori & Imperl, 2008). There are high demands for goalkeepers in terms of their performance, but they are often insufficiently trained since the performance limiting factors of the goalkeeper position are probably unknown (Kajtna et al., 2011) and training time is scarce (De Castro et al., 2011).
The aforementioned findings make it necessary to research positional differences in female handball players and the demands on goalkeepers in particular. The connection between position specialization and success has to be investigated. How goalkeepers' demand profiles differ from field players' needs to be specified. Furthermore, whether those differences (thought to indicate specialization) contribute to success (measured through an expertise index) also needs to be determined.
MATERIALS AND METHODS
Participants. 654 female players (18-52 years, 153-190 cm, 43-119 kg) of all German leagues from 1st Bundesliga to regional leagues were tested. Informed consent was obtained from all participants and the study complied with both the approval of the local medical research ethics committee and the current ethical standards of sports and exercise research (Helsinki Declaration).
Procedures. The players were tested using a test battery that assessed handball-relevant factors and involved conditional, constitutional, technical, tactical, psychological and biographical tests. All players and assistants were briefed before the tests. Questionnaires were distributed before testing so participants could complete them at home and bring them to the training site. The conditional, constitutional, technical and tactical profiles of the participants were recorded from May to September 2011 during their usual training.
Measurements. The test battery consisted of several tests covering a wide range of physical, psychological, technical and tactical factors relevant to handball (Table 1). In addition to two technique tests and a tactics test, there were also three psychological tests (standardized questionnaires). These covered most of the performance-relevant psychological factors: Volitional Components in Sports Questionnaire, Achievement-Motives Scale and Hakemp-Sport for action and state orientation in sports (Table 1). Also, a test battery for physiological factors was conducted (Table 1).
Body fat percentage (BF%) was measured using a caliper at the beginning of the participants' usual practice session. Calculations were done using body density (BD) and three skin folds (Siri,1956):
46 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Table 1
Test battery with references to studies where similar tests have already been used.
Test	Factor and testing procedure
5 x 20m Sprint (photo sensor DCT/F03, Cyclic velocity, endurance of velocity: Fastest and average time out of five attempts, Sportronic, Germany at start and finish, slow jog back to starting point and immediate start of next attempt.
Hulka & Belka, 2013)_
Jump & Reach (Moss, McWhannel, Jumping strength: Countermovement jump, jumping height measured as difference Michalsik & Twist, 2013)	between reaching height when standing and when jumping, accomplished height
marked with chalk on the players' fingers leaving marks on the wall, best of two attempts
Sit-ups (Hatzimanouil & Oxyzoglou, Endurance of strength (abdominal muscles): Maximum number of situps with feet on 2004)	a small box
Maximum number of chin-ups with Endurance of strength (arm muscles): Maximum number of chin-ups in angular supported heels	hanging with supported heels (regional-level players would not have been able to do a
(Büsch, Schorer & Lotz, 2008)	number of free chin-ups sufficient for calculation)
Reaction test with Basketball (Prätorius & Reaction speed: Participants have to stop a basketball rolling down a ramp within the Milani, 2008)	smallest possible rolling distance following an audio signal (mean of two attempts,
measured with a tape measure), standing with their back to the ramp at the lower end of the ramp, the ball being released from the upper end. Stand & Reach	Flexibility hamstrings / lower back: Reaching down to feet or beyond while standing
(Bös, 2001, Zapartidis et al.,	on a small box, distance between standing level and fingers measured with tape
2009 a, b)	measure, positive distance beyond feet,
negative above.
Throwing velocity with V-maxx throwing Elasticity arm muscles/throwing strength: Throwing from a standing position 5m in radar	front of the radar into the upper left corner of the goal, mean out of two attempts,
(EUROTronic technology, Germany) figuring that throwing speed is related to strength (van den Tillaar, 2004; Zapartidis et
_al., 2009 a, b)._
30m Sprint (Zapartidis et al., 2009 a, b) Cyclic velocity, start velocity, acceleration: Fastest and average out of two attempts with with splittimes at 5 and 10m	slow jog back to starting point and immediate start of next attempt.
Half Cooper Test* (6min. running, Bös, Basic endurance: Number of elliptic rounds (74m in length, marked with shuttles in the 2001)	training hall) within six minutes of running
Wall passing (Letzelter, Letzelter & Ball technique pass / catch: Time needed for 20 passes against a wall from a 4m
Scholl, 1988)_distance_
Slalom-dribbling with photo sensor DCT/ Ball technique Dribbling: 30m parcours, time measured with photo sensors at start and F03, Sportronic, Germany (Letzelter et finish
al., 1988)_
Tactics test via video (Wegner, Leptien & Tactical ability: IVS-video-test, 45 sequences, players have to solve match situations
Geode, 2010)	and receive points according to their answers
Skinfold measurement	Body fat percentage: Measurement of three skinfolds (Equation 1, 2)
(Whithers et al., 1987)_
Achievement Motives Scale, AMS (Elbe & AMS: hope for success and fear of failure (15 questions with 0-3 points each), net hope Wenhold, 2005)	(hope for success minus fear of failure) and total achievement motive (sum of hope for
success and fear of failure)
Volitional Components Questionnaire, VCQ: 0 to 3 points per question: for self-optimization (29 questions), self-impediment VCQ (Wenhold, Elbe & Beckmann, 2009) (9 questions), lack of activation (13 questions) and loss of focus (9 questions) Hakemp-Sport, Action/state orientation Hakemp-Sport: action/state orientation after malperformance, while planning a task in sports, HOSP (Beckmann & Wenhold, and while performing a task, 12 questions each with 0 to 1 points per question
2009)_
Players biography, leagues played each Players biography: Expertise points were given for each year played on a scale from 0 to past year	12; also, players were asked for age, body height and weight.
* Half Cooper test was used due to restricted time available in the training halls.
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 47
BD = 1,18562 - 0,08258*lg (LxTriceps, x_Subscapular, x_calf) [mm] Equation 1: Body density (Whithers et al., 1987). BF% = (4,95 / BD) - 4,5
Equation 2: Body fat percentage (Whithers et al., 1987).
When investigating performance the focus should be on successful players to determine the demands of the game. Success can be measured via expertise, which can then be tested for correlations with handball performance factors or positional differences for handball-relevant factors. According to Schorer (2007), the following factors can be used to measure expertise: efficiency and outstanding performance, duration and reproduction of excellent performance, excellent performance not only by accident, expertise through experience (ten-year rule, Ericsson & Lehmann, 1996), time spent training, long preparation, striving for excellence and perfection, motivation, and competition experience. Ericsson and Lehmann (1996) postulate that expertise is best measured quantitatively through competition performance. In this study, expertise was measured by assessing the leagues a player had competed in using data from a biographical questionnaire (in accordance with Sinuany-Stern, 1988, who used national leagues to measure expertise). Participants were asked to name all the clubs they had played for during their career to calculate an individual's expertise index on a scale of zero to twelve. International experience was classified as twelve points, 1st division as eleven points, down to only training, no competitions" as zero points.
The expertise index was calculated from the mean of nine expertise-influencing elements set forth in the literature: mean value of expertise points from the leagues played at senior level, expertise points in the highest league played, mean value of leagues of current and previous seasons, points in most frequent league, total playing experience overall in years scaled to 12 points1, points for highest league played during youth, mean value of all leagues played during youth, sum of expertise points at senior level (scaled as explained above), and sum of expertise points during youth scaled to twelve (see above).
The tested factors were measured in order to calculate specialization with a customised formula. This involved calculating whether each player differed more from all players of the same position or from all other players:
(x ,,, - value. ,, , , .,..)- (x , ... , - value. t . , , ... ,)
v alipiayers	factor1 player1 positionV v piayersposition1	factor1 player1 position^
Formula 1: Specialization on a position.
Statistical Analyses. Oneway ANOVAs (1x5: positions) per level were performed to investigate the variables in detail. The level of significance was set at p<0.05 and by trend-significance at p<0.10. Effect size was evaluated with r|2 (ETA partial squared), where 0.01<r|2<0.06 constitutes a small effect, 0.06<r|2<0.14 constitutes a medium effect and r|2>0.14 constitutes a large effect (Cohen, 1988).
The leagues that players competed in, sorted into competition levels (elite, subelite, regional), will be used as control factors as they could confound the results (Bortz & Dohring, 2006). Thus, it will be possible to gain detailed knowledge on positional differences within these levels without the
'12 points being the value of the best player who had played the most years or reached the highest number of points, all other players scaled accordingly
48 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
results being confounded by performance level. The main aim is to test for positional differences in the whole sample and in the different levels as well as interactions between players' positions and performance levels, operationalized.
Differences between levels are expected since the tested factors are relevant for performance in female team handball. If the factors are relevant to performance, players of higher leagues will have a more fitting skill level regarding psychological skill than players of lower leagues (in accordance with Letzelter et al., 1988). Therefore, differences between levels will only be tested to prove the validity of the tests.
Correlations between each player's expertise and their values for the different performance factors were calculated via Pearson's correlation coefficient, Spearman's rho and Kendall's tau b. Testing for correlations between the calculated differences (Formula 1) and the expertise of the different playing levels were also conducted with correlation levels of >0.1 (weak), >0.3 (moderate) and >0.5 (strong) and a confidence-interval of 0.95 (calculated according to Rinne, 2008). Linear regression will be used to create demand profiles for each position.
When calculating whether the ANOVA differed significantly between the leagues for the tested values, it is possible to determine which factors are important to goalkeepers (being more fittingly developed at the elite level, Letzelter, Letzelter & Scholl, 1988).
Experimental approach to the problem.
Hypotheses:
-	goalkeepers differ from field players regarding handball-relevant factors, to be tested via ANOVA;
-	goalkeepers are specialized for some factors, revealed through correlations between specialization and expertise (Pearson, Spearman's Rho, Kendall's Tau b);
-	their expertise will correlate with some factors usually seen as limiting factors for handball performance, but not all factors, to be tested via correlations between factor values and expertise
-	different factors than expected in literature will differ clearly between leagues and therefore be handball-relevant, to be tested via ANOVA
-	there will be a significant set of demands for the goalkeeper position.
The Statistical analysis was performed in SPSS, version 21.0 (SPSS, Inc., Chicago, IL).
RESULTS
A Oneway ANOVA with Scheffe post hoc testing (Table 2, 3) shows that goalkeepers have higher body weights and fat percentages than players of all field positions except pivots at all performance levels. They are taller than all field players except half backs. Goalkeepers are older than wing players at global level. For several psychological factors (Fear of failure, self-impediment, lack of activation, action/state orientation after malperformance) they display values higher than those of field position players at subelite and regional level (Table 2, 3). For most conditional factors, goalkeepers performed worse than field players. Only in the fastest attempt out of two for the 30 m sprint did they outperform all field positions except wings at elite and subelite level (see Table 2, 3).
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 49
Table 2
Descriptive statistics for anthropometric and psychological factors (Mean ± SD) and differences per competition level (Oneway ANOVA / Scheffepost hoc), only factors with significant differences to afield, position are shown.
Factors**	Effect n	Wingplayers (n = 197)		Half backs (n = 170)		Centre backs (n = 102)		Pivots (n = 92)		Goalkeepers (n = 91)	
Body height [cm]	0.436	168.10 ±	5.53 b c d e	174.94 ±	: 5.18 a c d	171.19	± 5.79 a b	170.84 ±	6.13 a b e	173.57	± 5.74 a d
Elite	0.564	167.89 ±	5.12 b c d e	177.23 ±	4.50 a c d	172.94 :	± 6.42 a b	172.71 ±	5.11 a b	176.18	± 5.56 a
Subelite	0.411	168.77 :	± 5.66 b e	175.18 ±	4.87 a c d	171.17	± 5.20 b	171.05	± 7.11 b	172.36	± 5.39 a
Regional	0.432	166.54 :	± 5.55 b e	171.86 :	± 5.22 a c	167.14	± 3.80 b	168.20	± 4.27	171.83	± 5.69 a
Body fat [%]	0.247	23.13 ±	: 3.63 d e	23.95	3.51 e	23.09 ±	3.38 d e	25.02 ±	3.31 a c	25.54 ±	3.93 a b c
Elite	0.346	21.87 ±	: 2.73 d e	22.16	3.46 d	22.65	± 3.13	24.92 ±	3.00 a b	24.25	± 3.42 a
Subelite	0.237	23.75	± 3.70 e	24.51	± 3.02	23.49	± 3.39 e	25.00	± 3.02	25.97 ±	3.57 a c
Regional	0.802	23.31	± 4.12 e	24.64	3.91 e	22.92	± 3.87	25.19 ±	4.13 e	26.76 ±	5.48 abd
Body weight [kg]	0.440	62.80 ±	7 12 b c d e	71.67 ±	9.00 a c	66.82 ±	6.97 a b d e	70.71 ±	8.64 a c	74.28 ±	12.18 a c
Elite	0.528	63.03 ±	7 21 b c d e	72.53	t 7.18 a	68.91	± 7.30 a	74.08 ±	6.56 a	73.12	± 6.65 a
Subelite	0.424	63.31 ±	7.27 b d e	71.36 ±	8.92 a c	65.83 ±	6.61 b e	68.46 ±	7.15 a e	74.21 ±	12.57 a c d
Regional	0.459	61.10 ±	6.53 b d e	71.43 ±	11.05 a	64.63	± 6.39	71.68 ±	11.84 a	77.27 ±	19.65 a
Age [a]	0.161	23.63 ±	5.80 d e	23.72	± 6.59	25.04	± 6.96	26.18 ±	7.33 a	26.08	± 7.45 a
Elite	0.298	21.49	± 3.81	20.90	± 3.67 c	24.24	± 6.08 b	24.04	± 4.98	23.79	± 4.69
Hope for success		34.48	± 5.32	33.11	± 6.25	33.34	± 5.66	33.60	± 6.56	32.90	± 7.32
Elite		35.64	± 4.99	33.95	± 5.41	34.21	± 6.81	33.50	± 7.75	33.63	± 6.64
Fear of Failure		10.85	± 7.43	11.86	± 8.01	10.15	± 6.65	9.98 ±	7.46	11.37	± 9.29
Elite		12.48	± 7.92	12.15	± 7.28	8.88	± 6.47	9.42 ±	7.16	10.26	± 5.98
Subelite	0.610*	8.44	± 7.58	9.58 ±	5.36 e	10.30 ±	: 5.17 e*	10.56	± 8.49	23.33 ±	17.04 b c*
Regional		8.65 ±	5.99 e	18.33	± 7.34	9.25	± 4.57	13.33	± 5.51	1.00 ±	0.00 a
Net Hope		23.56	± 10.78	21.25 :	± 12.13	23.20	± 10.38	23.73 ±	: 11.50	21.99	± 3.92
Elite		23.28	± 11.23	21.80 :	± 11.09	25.36	± 11.86	24.08 ±	: 13.31	24.78	± 9.55
Regional	0.470	27.12	± 7.70 e	22.58 ±	10.18 e	23.20 ±	: 8.39 e*	22.00	± 9.54	4.66 ± 14.74 a b* c*	
Total Achievement Motive		45.12	± 7.06	44.92	± 7.57	43.51	± 6.67	43.37	± 7.35	44.30	± 9.15
Elite	0.285	47.89 ±	6.47 c d*	46.10	± 6.45	43.12	± 5.88 a	42.92 ±	6.74 a*	44.54	± 7.70
Self-optimization		61.71	± 10.58	69.52	± 11.09	61.59	± 12.13	61.19 ±	: 12.19	59.88	± 11.80
Elite		63.87	± 9.53	60.24	± 10.89	64.91	± 13.34	59.13 ±	: 15.23	61.19	± 1.41
Self-impediment		11.23	± 4.34	11.85	± 5.09	10.90	± 4.29	10.53	± 4.52	10.88	± 5.11
Elite		12.70	± 4.69	12.85	± 4.65	10.72	± 4.33	12.13	± 3.85	11.59	± 4.50
Subelite	0.415	10.41	± 3.74	13.25 ±	5.43 c* e	7.71 ±	2.56 b	9.82 ±	3.66	8.46 ±	4.14 b
Lack of activation	0.145	8.36 ±	5.44 e	10.08	± 6.18	9.36	± 6.55	8.68 ±	6.59	10.93	± 7.07 a
Elite		7.06	± 5.20	10.32	± 6.56	8.78	± 6.64	10.04	± 8.50	9.42	± 5.58
Regional	0.381	10.11	± 6.13 e	11.33	6.57 e	9.13 ±	5.55 e	9.63 ±	5.07 e	18.58 ±	9.66 a b c d
Loss of focus		5.12	± 4.02	5.65	± 4.64	4.80	± 4.78	4.38 ±	3.58	5.22	± 3.95
Elite		4.45	± 3.79	5.76	± 4.93	3.81	± 4.40	5.00 ±	3.68	4.15	± 2.96
ASO alter malperformance		5.28	± 3.01	5.31	± 3.16	4.95	± 2.87	4.98 ±	2.76	5.94	± 3.18
Elite	0.277	4.03	± 2.88	4.27	± 3.07	5.58	± 3.20	3.73 ±	2.74	6.05	± 3.10
Subelite	0.496	5.29	± 3.65	4.41 ±	2.57 e	4.86	± 1.95	3.33 ±	3.23 e	7.69 ±	2.46 b d
Regional	0.409	5.92	± 3.57	5.90	± 2.02	2.40 ±	3.78 e	4.71	± 2.98	8.80	3.03 c
ASO when planning task		6.99	± 2.57	6.59	± 2.53	6.62	± 2.32	6.83 ±	2.32	6.56	± 2.79
Elite		6.96	± 2.49	6.59	± 2.77	7.12	± 2.32	6.83 ±	2.53	6.63	± 2.50
ASO when performing	0.122*	9.06 ±	: 1.95b	8.45 ±	2.52 a	8.79	± 2.22	9.01 ±	2.07	8.44	± 2.44
Elite	0.381	9.09	± 2.02	7.79 ±	2.35 d	9.26	± 1.73	10.43 ±	1.16 b	8.72	± 2.30
*by trend; ** Only levels with differenres to goalktepers arr shown in full detail, if no differences occurred, only global and elite level are shown.
a significant difference to wing players b significant difference to half backs c significant difference to centre backs d significant difference to pivots e significant difference to goalkeepers.
50 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Table 3
Descriptive statistics for conditional, technical and tactical factors (Mean ± SD) and differences per competition level (Oneway ANOVA / Scheffepost hoc), only factors with significant differences to a fie ld position are s ho con.
Factors **	Effect n	Wingplayers (n = 197)	Half backs (n = 170)	Centre backs (n = 102)	Pivots (n = 92)	Goalkeepers (n = 91)
Half Cooper Test [m] Elite Subelite Regional	0.207 0.324 0.210 0.341	1184± 135e 1288 ± 87 e 1174 ±106 1078 ±146 e	1177±129e 1265 ±108 e 1183±114e 1073 ±104 e	1207±111 e 1267 ± 95 e 1189 ±107 e 1130±92e	1164±141 1230 ± 113 1167±125 1083 ±157 e	1109 ± 149 abc 1184 ± 120 abc 1113 ± 103 bc 933 ± 177 abcd
Situps Elite		33 ± 19 45 ± 18	33 ± 21 46 ± 24	32 ± 20 38 ± 22	30 ± 16 39 ± 16	31 ± 21 42 ± 25
Chin-ups Elite Subelite	0.224 0.298 0.228	16 ± 7 e 20 ± 6 e 15 ± 7 e	15 ± 7 e 19 ± 9 13 ± 6	17 ± 7 e 19 ± 7 e 16 ± 6 e	14 ± 7 16 ± 6 14 ± 6	12 ± 6 abc 14 ± 6 ac 12 ± 6 ac
Throwing speed [km/h] Elite Subelite	0.253 0.445 0.268	58 ± 7 b 63 ± 6 b 57 ± 6 b	61 ± 8 ade 69 ± 6 acde 60 ± 7 ae	60 ± 6 64 ± 5 b 60 ± 6	57 ± 7 b 62 ± 5 b 58 ± 6	57 ± 7 b 61 ± 6 b 55 ± 6 b
Reaction-distance [m] Elite		0.82 ± 0.15 0.75 ± 0.14	0.80 ± 0.16 0.75 ± 0.19	0.78 ± 0.15 0.71 ± 0.12	0.80 ± 0.17 0.76 ± 0.16	0.82 ± 0.16 0.78 ± 0.17
Jump & Reach [m] Elite Subelite	0.205 0.217* 0.241	0.43 ± 0.06 0.45 ± 0.06 0.43 ± 0.06	0.45 ± 0.07 0.47 ± 0.06 de 0.46 ± 0.06 e	0.44 ± 0.06 0.43 ± 0.07 0.45 ± 0.06	0.42 ± 0.06 0.43 ± 0.07 b 0.42 ± 0.05	0.42 ± 0.07 0.43 ± 0.08 b 0.41 ± 0.07 b
Stand & Reach [m] Elite	0.126 0.795*	0.06 ± 0.01 0.07 ± 0.10	0.06 ± 0.01 0.07 ± 0.09	0.09 ± 0.01 0.10 ± 0.07	0.08 ± 0.01 0.07 ± 0.10	0.09 ± 0.01 0.09 ± 0.07
20m Minimum of 5 [s] Elite Subelite Regional	0.277 0.293 0.365 0.422	3.60 ± 0.25 de 3.44 ± 0.24 e 3.63 ± 0.22 e 3.76 ± 0.23 e	3.62 ± 0.24 de 3.44 ± 0.21 e 3.62 ± 0.19 e 3.79 ± 0.22 e	3.61 ± 0.22 e 3.51 ± 0.15 3.63 ± 0.22 e 3.76 ± 0.24 e	3.73	± 0.29 ab 3.57 ± 0.31 3.74	± 0.19 3.90 ± 0.32 e	3.82 ± 0.39 abc 3.62 ± 0.28 ab 3.85 ± 0.25 abc 4.29 ± 0.73 abcd
20m Mean of 5 [s] Elite Subelite Regional	0.274 0.359 0.348 0.448	3.72 ± 0.27 e 3.49 ± 0.17 e 3.75 ± 0.23 e 3.92 ± 0.27 e	3.73 ± 0.26 e 3.51 ± 0.21 e 374 ± 0.21 e 3.94 ± 0.23 e	3.73 ± 0.25 e 3.59 ± 0.16 3.77 ± 0.26 e 3.90 ± 0.26 e	3.83 ± 0.29 3.61 ± 0.26 3.87 ± 0.22 4.00 ± 0.30 e	3.96 ± 0.41 abc 3.72 ± 0.26 ab 3.98 ± 0.26 abc 4.51 ± 0.73 abcd
30m Minimum of 2 [s] Elite Subelite	0.141 0.459 0.473	5.03 ± 0.39 4.83 ± 0.24 bcd 5.05 ± 0.31 b	5.09 ± 0.44 5.24 ± 0.36 ae 5.00 ± 0.52 acde	5.15 ± 0.42 4.92 ± 0.21 ae 5.07 ± 0.30 b	5.14 ± 0.41 5.06 ± 0.37 a 5.10 ± 0.34 b	4.98 ± 0.30 4.87 ± 0.22 bc 5.05 ± 0.29 b
30m Mean of 2 [s] Elite Subelite Regional	0.293 0.338 0.389 0.455	5.03 ± 0.33 e 4.82 ± 0.20 e 5.09 ± 0.32 e 5.23 ± 0.33 e	5.07 ± 0.40 e 4.83 ± 0.25 e 5.07 ± 0.27 e 5.43 ± 0.57 e	5.02 ± 0.29 e 5.09 ± 0.33 5.012 ± 0.55 e 5.19 ± 0.38 e	5.20 ± 0.43 4.98 ± 0.39 5.20 ± 0.33 e 5.52 ± 0.48	5.38 ± 0.54 abc 5.07 ± 0.31 ab 5.43 ± 0.37 abcd 6.17 ± 0.99 abc
Wall-passing [s] Elite Regional	0.179 0.214* 0.326	26.51 ± 2.69 24.84 ± 1.85 29.10 ± 3.02	26.20 ± 2.49 e 24.55 ± 2.09 27.89 ± 2.72 e	25.80 ± 2.34 e 24.88 ± 2.21 27.55 ± 2.36 e	26.52 ± 2.60 24.97 ± 1.63 28.51 ± 3.18	27.41 ± 3.01 b c 25.94 ± 2.41 31.04 ± 2.93 b c
Slalom-dribbling [s] Elite Subelite Regional	0.261 0.366 0.361 0.367	7.60 ± 0.56 e 7.16 ± 0.30 e 7.66 ± 0.48 e 8.06 ± 0.56	7.56 ± 0.48 e 7.20 ± 0.33 e 7.56 ± 0.39 e 8.00 ± 0.44 e	7.46 ± 0.42 e 7.23 ± 0.31 e 7.45 ± 0.35 e 8.06 ± 0.28	7.79 ± 0.73 7.46 ± 0.71 7.71 ± 0.34 8.06 ± 0.56 e	7.95 ± 0.71 a b c 7.56 ± 0.41 a b c 7.99 ± 0.52 a b c 8.89 ± 1.24 b d
Tactics Elite		47.72 ± 8.87 50.74 ± 8.57	47.97 ± 9.68 53.46 ± 7.50	49.65 ± 8.99 51.61 ± 9.54	46.46 ± 9.34 50.20 ± 6.86	46.99 ± 9.84 49.32 ± 9.16
*by trend; ** Only levels with differences to goalkeepers are shown in full detail, if no differences occurred, only global and elite level are shown.
a significant difference to wing players b significant difference to half backs c significant difference to centre backs d significant difference to pivots e significant difference to goalkeepers.
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 51
Table 4
Test for performance relevance/separation between leagues (Oneway ANOVA), test for correlation between performance and team expertise, test for correlation between specialization and team expertise (s too ngest corre lation l evels out of Peterson coefficient, Spearman's rho, Kendall's tau b), descriptive statistics at e lite level (Mean ± SDj and modet values for goalkeepers in 3rtl league ((Calculated value ± Estimated error).
	Separation between leagues (Effect-size n)	Correlation level performance	Correlation level specialization	Values elite level (descriptive statistics)	Model goalkeeper 3rd league (linear regression)
Half Cooper Test	0.717	0.477	- 0.479	1184±102	1193±122
Situps	0.490	0.413	- 0.448	42 ± 25	38 ± 20
Chin-ups	0.259*			14 ± 6	13 ± 6
Throwing-speed	0.647	0.551	- 0.549	61 ± 6	60 ± 6
20m Minimum out of 5 [s]	0.705	- 0.332	- 0.332	3.62 ± 0.28	3.67 ± 0.34
20m average out of 5 [s]	0.740	- 0.365	- 0.365	3.72 ± 0.26	3.78 ± 0.34
30m Minimum out of 2 [s]	0.517	- 0.254	0.254	4.87 ± 0.22	4.92 ± 0.29
30m average out of 2 [s]	0.738	- 0.435	- 0.435	5.07 ± 0.31	5.13 ± 0.43
Jump & Reach [m]		0.198*	- 0.382	0.43 ± 0.08	0.43 ± 0.08
Stand & Reach [m]				0.09 ± 0.07	0.09 ± 0.07 ** ***
Reaction-distance [m]	0.327	- 0.264	- 0.274	0.78 ± 0.17	0.78 ± 0.16
Body height [m]	0.107	0.301	0.306	1.76 ± 0.06	1.76 ± 0.03
Body-fat percentage [%]	0.302			24.25 ± 3.42	24.41 ± 1.93
Body weight [kg]	0.519			73.12 ± 6.65	72.79 ± 6.10***
Age [a]	0.324	- 0.259	- 0.287	23.79 ± 4.69	22.86 ± 3.47
Wall-passing [s]	0.560	- 0.547	- 0.546	25.94 ± 2.41	25.50 ± 2.58
Slalom-dribbling [s}	0.761	- 0.530	- 0.530	7.58 ± 0.41	7.50 ± 0.62
Tactics-Test	0.547	0.309	- 0.319	49.32 ± 9.16	50.67 ± 9.41
Hope for success	0.336	0.213*	- 0.219	33.63 ± 6.64	34.65 ± 7.22
Fear of failure				10.26 ± 5.98	10.01 ± 9.28**
Net hope	0.362	0.232	- 0.245	24.78 ± 9.55	25.81 ±13.65
Total achievement motive				44.54 ± 7.70	45.09 ± 9.18** ***
Self-optimization			- 0.185*	61.19 ± 10.41	62.20 ± 11.71
Self-impediment	0.522	0.162	- 0.225	11.59 ± 4.50	11.64 ± 5.11**
Lack of activation	0.455	- 0.276	- 0.294	9.42 ± 5.58	7.88 ± 6.67
Loss of focus	0.354		- 0.196*	4.15 ± 2.95	4.04 ± 3.85
ASO after malperformance		- 0.325	- 0.320	6.04 ± 2.79	5.55 ± 3.19** ***
ASO while planning a task				6.63 ± 2.50	6.87 ± 2.79** ***
ASO while performing a task	0.474			8.56 ± 2.36	8.92 ± 2.42
ASO = Action /state orientation.
* by trend; ** p-value not sufficient; *** effect not sufficient. . no significant results.
Reliability is given with confidence-interval 0.95 (Rinne, 2008).
52 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
The Oneway ANOVA separated significantly or at least by trend between performance levels for all the tested performance factors except for jump & reach, stand & reach, fear of failure, total achievement motive, self-optimization and action/state orientation after malperformance and when planning a task (Table 4).
There are strong, significant correlations between performance and expertise for throwing speed, wall passing and Slalom dribbling. Furthermore, there were moderate correlations between the two for the Half Cooper test, sit-ups, 20m sprint (fastest out of two attempts and the mean time of five runs), 30m sprint (mean out of two runs), body height, tactics skill and action control after malperformance. Finally, there were weak correlations between the two for 30m sprint (fastest time out of two), jump & reach (by trend), reaction distance, age, net hope, self-impediment and lack of activation.
The direction of the correlations should be noted, especially for self-impediment (Table 4). When considering specialization of the goalkeeper position, there are different directions for the correlations between specialization and expertise for each factor (Table 4). There were factors without correlations between specialization and expertise (Chin-ups, stand & reach, body height, body fat percentage, fear of failure, total achievement motive and action/state orientation when planning and performing a task), factors with positive correlations between specialization and expertise (30m fastest time and body height) and factors with negative correlations between specialization and expertise (all other factors, see Table 4).
All the factors, except for stand & reach, body weight, fear of failure, total achievement motive, self-impediment and action/state orientation after malperformance and while planning a task, had significant model values with sufficient effect sizes that could be calculated for German 3rd league goalkeepers (Table 4).
H J lib Jt* 5.
13%	CenUe Backs
(Confidence interval 0.95 with d_max ± 2.15 %)
WP = Wingplayers, HB = Half Backs, CB = Centre Backs, P = Pivots, GK = Goalkeepers
Figure 1: Percentages of handedness on different p ositions and in female population.
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 53
Goalkeepers differ from both field position players and the female population regarding handedness (Figure 1). The percentage of right-handers in a league correlates negatively with expertise (-0.993) while being ambidextrous correlates positively with expertise (0.858) for goalkeepers.
DISCUSSION AND CONCLUSION
As delineated above, goalkeepers show a significant negative difference from field players over a wide range of factors. These represent deficits in goalkeeper fitness and coaching of psychological factors in all sectors except tactics. With negative correlations for specialization and expertise (Table 4), the assumption must be made that there are training deficits in comparison to the field players or simply no demand for several performance factors in goal. Goalkeepers match or exceed some field positions only in body height (which must be considered an advantage in play) and fastest 30m sprint out of two attempts (conflicting with the results of Ignat'eva et al., 2002; Table 2, 3). Both this study and previous research (Ignat'eva et al., 2002; Zapartidis et al., 20011) found goalkeepers are less physically fit than field players for most factors.
The factors for which the Oneway ANOVA separated significantly or at least by trend between performance levels (see results) must be considered at least slightly relevant to the goalkeeper position. However, some of the effects occur only by trend or have a relatively low effect size. The only strong significant effects recorded were for chin ups, stand & reach, body weight, body fat percentage, total achievement motive, fear of failure, self-optimization, loss of focus and action/ state orientation when planning and performing a task (Table 4). Strong correlations can therefore only be found for technical factors and throwing speed. Consequently, the study suggests that several performance factors are of only moderate relevance for goalkeepers (Table 4).
When considering specialization for the goalkeeper position, specialization and expertise arise for the different factors, as reported in the results (Table 4). No correlation means that goalkeepers are not a homogenous group, which leads to doubts about the importance of those factors for the goalkeeper position. Positive correlations suggest the need for specialization. Negative correlations suggest the necessity of adjusting to meet the performance of the field positions for the concerned factors. As the factors that correlate negatively with specialization also correlate positively with expertise and differ significantly between leagues, this means that goalkeepers do not match the performance of the players on field positions (Table 2, 3).
Altogether the findings shown in table 4 suggest that goalkeeper core demands differ from those of field players across the leagues. However, considering the negative specializations in top teams with an EI of 7 or higher (Table 4), goalkeepers should evince a basic fitness and adapt to the level of their teammates to ensure a homogeneous group and adequate performance.
For several factors, significant model values with sufficient effect sizes could be calculated for players in German 3rd league players (Table 4). Therefore, the concerned factors might be relevant when wanting to play at elite level.
Goalkeepers did not achieve high values and only had a correlation by trend with expertise for the jump & reach test. Perhaps the rate at which force is developed when jumping is the relevant factor in goal and not the player's actual jumping height (as assumed by Pori et al., 2011).
Both coaches and the literature delineate flexibility as a core demand for all positions but especially for goalkeepers. However, to our knowledge this has not yet been confirmed by a research
54 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
study (Pori, Sibila, Justin, Kajtna & Pori, 2012; Rogulj et al., 2005). It must be said that results for stand & reach correlate with reaching height as measured for the jump & reach test and body height. At the subelite level, goalkeepers and centre backs are more flexible than players of other positions. This is congruent with the results of Zapartidis et al. (2011). The special status of goalkeepers in terms of flexibility as postulated by Rogulj et al. (2005) could therefore only be partially confirmed. This could possibly be caused by insufficient work with goalkeepers during training (as assumed by Zapartidis et al., 2009 a) or differences in static and dynamic flexibility. Also, a connection between flexibility and elasticity/springiness seems likely.
Literature considers goalkeepers to be relatively heavy (Sibila et al., 2008). Urban and Kandac (2011) found the heaviest female European Championship players to be pivots, followed by goalkeepers. Sibila, Pori and Imperl (2008) assumed goalkeepers had a higher body weight than some field positions. In the present study, goalkeepers were heavier than field players at low performance levels. Urban and Kandrac (2011, female European Championship players) found that pivots had the highest percentage of body fat followed by goalkeepers, centre backs, half backs and line players (in that order). In the present study, goalkeepers also displayed the second highest percentage of body fat after pivots. It seems to be necessary to avoid obesity in female goalkeepers in lower leagues. This is seen in the negative correlation between specialization and expertise. Goalkeeper selection during youth training is in some cases a result of high body weight (obesity), greater body height and inadequate motor ability of the "chosen" child, but is rarely caused by a child volunteering (Sibila, Pori & Imperl, 2008). The results of the current study also hint that players with training deficits are placed in goal to keep them and their constitutional, tactical, technical and conditional deficits out of the attack (Matthys, 2012). Differences between field players and goalkeepers in psychological aspects might be the result of the above-mentioned mechanisms. Goalkeepers might be de-motivated by these "selection" mechanisms. A more careful selection process as well as more attention and coaching is needed to improve this process.
Literature describes goalkeepers as tall (Cavala et al., 2013; Milanese et al., 2011; Zapartidis et al., 2009 a, b). Urban and Kandrac (2011) found female elite half back players to be the tallest, followed by goalkeepers. In the current study, goalkeepers displayed values above the average of all positions in the top teams and should perhaps be specialized in regards to body height.
It has been found that successful male goalkeepers spend less time thinking about failures than less successful goalkeepers (Kajtna et al., 2011). This can also be confirmed for female players, since action/state orientation after malperfomance correlates negatively with expertise.
In the present study, no relation can be found between reaction distance and expertise for goalkeepers, who had the slowest values of all the positions. It is possible that goalkeepers were unable to produce quick accelerations on the way to the measuring point in the reaction test. Less successful goalkeepers display faster reaction times, better reaction to simple stimuli and respond faster to simple visual stimuli (Kajtna et al., 2011). Thus, anticipation might be more important than reaction-speed (Schorer, 2007). When coaches demand expertise in the goal, anticipation expertise might be what is meant (Schorer, 2007). Goalkeepers were older than field players in the present study. In the top teams the average age for each of the field positions ranges between 20 and 27 years. Goalkeepers were aged up to 32 years in Bundesliga 1 to 3 and between 22 and 30 years in the top teams.
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 55
Ignat'eva et al. (2002) have found that female adult goalkeepers are conditionally weaker than field players. Some of the conditional and technical differences are only prevalent in the lower leagues and therefore might not be desired, since at the elite level there are goalkeepers who do match the field players' performance for those factors. This can also be seen in the correlation between specialization and expertise. The psychological demands for goalkeepers are not clear yet either (Kajtna et al., 2011). Recent test batteries have found contradictory or but few significant results and have therefore probably only identified a few of the core demands for goalkeepers (Kajtna et al., 2011; Pori, Sibila, Justin, Kajtna & Pori, 2012). Zapartidis et al. (2009 b) found that there was no significant difference between goalkeepers chosen for the national team and those not chosen.
Many of the tested factors are not crucially relevant for goalkeepers. A tangible demands profile is either nonexistent or very narrow. The demands of the goalkeeper position need to be tested using a test battery that takes position specific goalkeeper demands into account, as already suggested by Kajtna et al. (2011). Only a few core demands can be identified (Table 4), such as reactive force (considering present results together with the statement of Pori et al., 2011), experience ("anticipation expertise", Schorer, 2007), ambidexterity, low action/state orientation after malperformance, throwing speed and 30m sprint (fastest time out of two). To a certain extent, technical factors (passing, catching and dribbling) as well as sprints over 20 and 30m are also relevant, as are other factors linked to basic fitness (e. g., Half Cooper, reaction distance, sit-ups) and constitutional factors such as body height, age and tactical ability. Hope for success, net hope, self-impediment and lack of activation are also of minor importance. However, there may also be other as yet unidentified factors.
Positioning tall players in goal solely due to their body height is not recommended, as important field techniques cannot be easily acquired later (Matthys, 2012), making a possibly advantageous change of position almost impossible later in a player's career. Constitutional factors especially do not seem to be entirely adequate as a selection criterion during position-specific selection (Gon^alves et al., 2012 and Moesch, Hauge, Wikman & Elbe, 2013, both preferring psychological factors; Matthys, 2012; Visnapuu & Jürimäe, 2009: only sitting height correlates with motor parameters; Zapartidis et al., 2009 b). However, some authors see them as important predicators (Cavala et al., 2013).
Position-specific selection and training are related to performance and must be carried out under consideration of gender and specialization age. Position-specific selection should not be too heavily based on constitutional parameters below senior level (see also Matthys, 2012). The development of handball-relevant psychological factors during puberty has to be researched in order to determine whether they could be a better predicator for excellence in team handball than constitutional or conditional factors, especially for female handball goalkeepers. Psychological factors might play a role in talent selection, as biological factors have disadvantages when used as predicators.
REFERENCES
Beckmann, J. & Wenhold, F. (2009). Handlungsorientierung im Sport. [Action control in sports] Köln: Sportverlag Strauß.
Bös, K. (2001). Handbuch Motorische Tests. [Handbook motoric tests] Hogrefe: Göttingen.
56 Analysis of handball goalkeepers' performance
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Bortz, J. & Döhring, N. (2006). Forschungsmethoden und Evaluation für Human- und Geisteswissenschaftler [Scientific methods and evaluation for humanities and arts]; 4th, revised Edition, Heidelberg : Springer.
Büsch, D., Schorer, J. & Lotz, S. (2008). DHB-Tagung 2008: Vorläufige Auswertung der Sichtung 2008. [German Handball Federation Congress 2008: Preliminary evaluation of the inspection 2008] In cooperation with: IAT Leipzig, DHB and Westfalian Wilhelms-Universiy of Münster: Leipzig and Münster.
Cavala, M. & Katie, R. (2010). Morphological, Motor and Situation-Motor Characteristics of Elite Female Handball Players According to Playing Performance and Position. Collegium Anthropologicum, 34(4), 1355 - 1361. Cavala, M., Trninie, V., Jasie, D. & Tomljanovie, M. (2013). The Influence of Somatotpye Components and Personality Traits on the Playing Position and the Quality of Top Croatian Cadet Female Handball Players. Collegium Anthropologicum: 37(2013), 93 - 100.
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, New Jersey: L. Coren, S. (1993). The Left-Handed Syndrome. New York : Vintage Books.
De Castro, J. A. P. R., de Sequeira, P. J. R. M. & Cruz, C. M. C. (2011). Goalkeeper - Specific training for youngsters : characterization of the importance and structure in the formation process of handball goalkeeper. In: EHF, ed. EHF Scientific Conference 2011. Haugsdorf: Hofer, 2011. pp. 238 - 242.
Elbe, A.-M., & Wenhold, F. (2005). Cross-Cultural Test Control Criteria for the AMS-Sport. International Journal of Sport and Exercise Psychology, 3, 163-178.
Ericsson, K. & Lehmann, A. C. (1996). Expert and exceptional performance: Evidence on maximal adaptations on task constraints. Annual Review of Psychology, 47, 273 - 305.
Gon^alves, C. E. B. , Rama, L. M. L. & Figueiredo, A. B. (2012). Talent Identification in Sport: An Overview of Some Unanswered Questions. International Journal of Sports Physiology and Performance, 2012 (7), 390 - 393. Hatzimanouil, D., Oxyzoglou, N. (2004). Evaluation of the morphological characteristics and motor skills in the national junior handball teams of Greece and Yugoslavia, Journal of Human Movement Studies, 46 (2), 125 -140. Hulka, K. & Belka, J. (2013). Design and validity of specific handball related sprint ability test. In EHF (Ed.), In EHF (Ed.), Proceedings of the 2nd EHF-conference 2013 in Vienna, Haugsdorf: Hofer, pp. 63 - 66).
Ignat'eva, V. & Minabutdinov, R. (2014). Test tasks to control physical fitness of elite handball players. Theory and practice of physical culture, (4), 15.
Ignat'eva, V. J., Petracheva, I. V. & Savinkov, E. O. (2002). Specifications of Physical Preparedness of Elite Female Handball Players of Various Game Roles. Russian State Academy of Physical Culture, Moscow.
Kajtna, T, Pori, M, Justin, I. & Pori, P. (2011). Psychological characteristics of Slovene handball goalkeepers. In: EHF, ed. EHF Scientific Conference 2011. Haugsdorf: Hofer, 2011. pp. 73 - 77.
Karcher, C. & M. Bucheit (2014). On-Court Demands of Elite Handball, with Special Reference to Playing Positions. Sport Medicine, 44(6), 797-814.
Letzelter, H, Letzelter, M & Scholl, H. (1988). Methodologische Probleme in der Sportspielforschung (1st ed.). [Methodologic problems in sports games research] Hamburg: Czwalina.
Manchado, C., Tortosa, J., Vila, H., Ferragut, C. & Platen, P. (2013). Performance Factors in Women's Team Handball and Physiological Aspects - A Review. The Journal of Strength and Conditioning Research, 27 (6), 1708 - 1719. Marczinka, Z. (2011). What's the Difference? - Coaching Female and Male Handball Players. In: EHF (Ed.) EHF Scientific Conference 2011. Haugsdorf: Hofer (pp. 89 - 93).
Matthys, S. (2012). Talent identification, development and selection in youth handball players : contribution of cross-selectional and longitudinal measures of anthropometry, physical performance and maturation. Dissertation, University of Ghent.
Milanese, C, Piscitelli, F, Lampis, C, & Zancanaro, C. (2011). Anthropometry and body composition of female handball players according to competition level or the playing position. Journal of Sports Sciences, 29 (12), 1301 - 1309. Moesch, K., Hauge, M-L. T., Wikman, J. M., & Elbe, A-M. (2013). Making it to the top in team sports: Start later, intensify, and be determined. Talent Development and Excellence, 5 (2), 85-100.
Moss, S., McWhannel, N., Michalsik, L B. & Twist, C. (2013). Anthropometric and physical performance characteristics of elite and non-elite youth female handball players. In EHF (Ed.), Proceedings of the 2nd EHF-conference 2013 in Vienna, Haugsdorf: Hofer, pp. 96 - 101).
Kinesiologia Slovenica, 24, 1, 43-57 (2018)
Analysis of handball goalkeepers' performance 57
Pori, M, Justin, I, Kajtna, T & Pori, P. (2011). Which Motor Abilities have the Highest Impact on Competition Performance of Slovenian Handball Goalkeepers? In: EHF, ed. EHF Scientific Conference 2011. Haugsdorf: Hofer. pp. 294 - 298.
Pori, P, Šibila, M, Justin., I, Kajtna, T & Pori, M. (2012). Correlation between the Motor Abilities and Competition Performance of Slovenian Handball Goalkeepers. Kinesiologia Slovenica, 18 (2), 19 - 26.
Prätorius, B. & Milani, T. L. (2008). Entwicklung eines Koordinationstests für Kinder im Grundschulalter und dessen Validierung mit Hilfe biomechanischer Methoden. [Development of a coordinational test for children in grammar school and ist validation with biomechanical methods] Göttingen: Cuvillier.
Rinne, H. (2008). Taschenbuch der Statistik [Handbook statistics] (4th revised ed.). Frankfurt a. M.: Verlag Harri Deutsch.
Rogulj, N, Srhoj, V, Nazor, M, Srhoj, L & Čavala, M. (2005). Some Anthropologic Characteristics of Elite Female Handball Players at Different Playing Positions. Collegium Anthropologicum, 29 (2), 705-709.
Schorer, J. (2007). Höchstleistung im Handballtor. [Maximum performance in the handball goal] Dissertation, Ruprecht-Karls-University of Heidelberg.
Siri, W. E. (1956). The gross composition o the body. In: Tobias, C. A., Lawrence, J. H. (Eds.): Advances in biological and medical physics Vol. 4, p. 239. Academic, New York.
Šibila, M, Pori, P, & Imperl, D. (2008). Rokometni vratar: tehnika, taktika, metodika [Handball goalkeeper: Techniques, tactics, methods]. Ljubljana: Fakulteta za šport.
Speicher, U., Kleinöder, H., Klein, G. D., Schack, T. & Mester, J. (2006) Eine Analyse der kognitiven Handlungsschnelligkeit von Handballtorhüterinnen als Basis für eine effektive Trainingssteuerung. Leistungssport, 36 (6), 11 - 15. [An analysis of cognitive speed of action of female handball goalkeepers as basis of effective training control]
Urban, F. & Kandrač, R. (2011). The Relationship between Morphological Profile and Player Performance in Elite Female Handball Players. www.researchgate.net, 4.4.2014.
Urban, F, Kandrač, R & Taborsky, F. (2011 a). Position-specific Anthropometric Profiles: 2011 Women's 19th European Handball Championship. www.researchgate.net.
Visnapuu, M & Jürimäe, T. (2009). Relations of anthropometric parameters with scores on basic and specific motor tasks in young handball players. Perceptual and Motor Skills, 108, 670 - 676.
Wagner, H, Finkenzeller, T., Würth, S. & Serge P. von Duvillard (2014) Individual and Team Performance in Team-Handball: A Review Article. Journal of sports science & medicine, 13, 808-816.
Weber, J. & Wegner, M. (2016 a). Constitutional demands for different playing positions in female team handball, Sportwissenschaft, 46 (4), 305 - 314.
Weber, J. & Wegner, M. (2016 b). Are there different psychological profiles per playing position in female team handball? Talent Development & Excellence, 8 (2), 52 - 63.
Wegner, M, Leptien, L & Goede, T. (2010). Virtuelle und reale Entscheidungssituationen im Handball - ein Leistungsmerkmal?!. [Virtual and real decision-making in handball - determining performance?!] In: Proceedings of the dvs-Sportspielsymposium 2010, Münster.
Wenhold, F, Elbe, AM & Beckmann, J. (2009). Volitionale Komponenten im Sport. [Volitional components in sports] Köln: Sportverlag Strauß.
Zapartidis, I., Toganidis, T., Vareltzis, I. & Christodoulidis, T. (2009 a). Profile of you female Handball players by playing position. Serbian Journal of Sports Sciences, 3 (1-4), 53 - 60.
Zapartidis, I., Vareltzis, I., Gouvali, M. & Kororos, P. (2009 b). Physical fitness and anthropometric characteristics in different levels of young team handball players. The Open Sports Sciences Journal, 2, 22 - 28.
Zapartidis, I., Kororos, P., Christodoulidis, T., Skoufas, D. & Bayios, I. (2011). Profile of young handball players by playing position and determinants of ball throwing velocity. Journal of Human Kinetics, 27, 17 - 30.