ArticlePDF Available

Climbing anxiety scale (CAS-20): Preliminary development and validation

April 2024
Psychology of Sport and Exercise 73(1):102635

DOI:10.1016/j.psychsport.2024.102635

Authors:

Maria Stefania Ionel

Babeş-Bolyai University

Andrei Ion

University of Bucharest

Dragos Iliescu

University of Bucharest

Anxiety has been the primary focus of emotion research in sport psychology. Most of the existing anxiety measures focus on the competition related anxiety. Little is known about the way in which anxiety affects athletic outcomes in extreme sports. We contribute to the literature on anxiety in extreme sports by: (1) developing and providing a preliminary validation for a novel, theoretically anchored sport climbing inventory, Climbing Anxiety Scale (CAS-20), among an international sample of rock-climbers (N = 153); and (2) providing preliminary evidence on its factorial and criterion-related validity. Our investigation includes two phases. The first phase (6 clinical and sport psychology experts) included the development and expert review of a climbing specific anxiety scale. The second phase (N = 153) offers preliminary evidence pertaining to the measure’s reliability, factorial, convergent and criterion related validity. Factorial validity was investigated by deploying a series of confirmatory factorial analyses. Convergent and discriminatory validity were examined by comparing the scale’s associations with a general anxiety measure, a sport anxiety measure, as well as climbing self-efficacy. Criterion-related validity was estimated by examining its relationship with rock-climbing performance. We contribute to the general domain of sport and athletic research by developing a sport-specific anxiety measure, investigating whether and how anxiety comes into play in rock-climbing, a high-risk sport. This scale can be used for assessing anxiety in climbing and monitoring the impact of an interventions designed to reduce these symptoms.

Content uploaded by Maria Stefania Ionel

Content may be subject to copyright.

CORRECTED PROOF

Psychology of Sport & Exercise xxx (xxxx) 102635

Contents lists available at ScienceDirect

Psychology of Sport & Exercise

journal homepage: www.elsevier.com/locate/psychsport

Climbing anxiety scale (CAS-20): Preliminary development and validation

Maria Stefania Ionel a, Andrei Ion b,⁎, Dragos Iliescu b,c, Laura Visu-Petra a

aResearch in Individual Differences and Legal Psychology –RIDDLE Lab, Department of Psychology, Faculty of Psychology and Educational Sciences, Babes-Bolyai University,

Cluj-Napoca, Romania

bAssessment and Individual Differences –AID Lab, Department of Psychology and Cognitive Science, University of Bucharest, Romania

cStellenbosch University, South Africa

ARTICLE INFO

Keywords:

Sport anxiety

Measurement

Scale development

Validation

Rock-climbing

ABSTRACT

Anxiety has been the primary focus of emotion research in sport psychology. Most of the existing anxiety mea-

sures focus on the competition related anxiety. Little is known about the way in which anxiety affects athletic

outcomes in extreme sports. We contribute to the literature on anxiety in extreme sports by: (1) developing and

providing a preliminary validation for a novel, theoretically anchored sport climbing inventory, Climbing Anxi-

ety Scale (CAS-20), among an international sample of rock-climbers (N= 153); and (2) providing preliminary

evidence on its factorial and criterion-related validity. Our investigation includes two phases. The first phase (6

clinical and sport psychology experts) included the development and expert review of a climbing specific anxiety

scale. The second phase (N= 153) offers preliminary evidence pertaining to the measure’s reliability, factorial,

convergent and criterion related validity. Factorial validity was investigated by deploying a series of confirma-

tory factorial analyses. Convergent and discriminatory validity were examined by comparing the scale’s associa-

tions with a general anxiety measure, a sport anxiety measure, as well as climbing self-efficacy. Criterion-related

validity was estimated by examining its relationship with rock-climbing performance. We contribute to the gen-

eral domain of sport and athletic research by developing a sport-specific anxiety measure, investigating whether

and how anxiety comes into play in rock-climbing, a high-risk sport. This scale can be used for assessing anxiety

in climbing and monitoring the impact of an interventions designed to reduce these symptoms.

1. Anxiety in sports and athletic performance

The study of anxiety and its impact on sport-relevant outcomes has a

long history, anxiety being one of the most commonly measured con-

structs in competitive sports (Cox et al., 2003). Furthermore, the preva-

lence of anxiety symptoms varies, with rates ranging from 26% among

former elite athletes to 34% among current elite athletes (Gouttebarge

et al., 2019), and reaching up to 45% in team sports (Reardon et al.,

2019) depending on the type of anxiety and the measures used to cap-

ture it, the most common being clinical interviews and self-report mea-

sures (Reardon et al., 2019). Extant research proposed various path-

ways through which anxiety can negatively impact athletic perfor-

mance, such as physiological activation, kinematic variables, or atten-

tional impairment (Cooke et al., 2011;Mullen et al., 2005). Considering

both the theoretical foundations (Liebert & Morris, 1967) and the com-

plexity of the pathways linking anxiety symptoms to athletic outcomes,

multifaceted measures of anxiety in sports have been developed (e.g.,

Jones et al., 2019). Although much progress has been made to under-

stand the intricacies of the anxiety-athletic outcome connection, little is

known about the way in which this relationship unfolds in high-risk

sports. To this end, we developed a climbing anxiety measure, provid-

ing preliminary evidence pertaining to its validity and exploring its as-

sociation with athletic performance in rock-climbing.

2. Conceptualizing anxiety in sport

Anxiety has been at the forefront of research aimed at understand-

ing the way in which emotions facilitate or interfere with athletic out-

comes and athletic behaviours. Anxiety received the lion’s share of at-

tention in sport psychology with several theoretical explanations ac-

counting for its effects (Janelle et al., 2020). Despite this extensive the-

oretical development, the number of sport-specific measures aiming at

capturing it is not so high. This discrepancy is more pronounced when it

comes to non-competitive sports, with only a few notable exceptions,

such as the study conducted by Llewellyn et al. (2008). Depending on

the construct level, these measures are structured with either a unidi-

⁎Corresponding author. Department of Psychology, University of Bucharest, 90th Panduri Ave., Bucharest, 030018, Romania.

E-mail address: andrei.ion@fpse.unibuc.ro (A. Ion).

https://doi.org/10.1016/j.psychsport.2024.102635

Received 29 September 2023; Received in revised form 27 February 2024; Accepted 2 April 2024

1469-0292/© 20XX

Note: Low-resolution images were used to create this PDF. The original images will be used in the final composition.

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

mensional or multidimensional approach, with most of the measures fo-

cusing on competitions, and assessing state and trait anxiety. One ap-

proach conceptualises anxiety in sport as a unidimensional construct,

without differentiating between various sub-components such as so-

matic or cognitive anxiety. Within this approach, one of the widely used

measures is the Sport Competition Anxiety Test (SCAT; Martens, 1977).

Another approach describes anxiety in sport as being a multi-faceted

construct. For example, Martens et al.'s (1990) multidimensional anxi-

ety theory became one of the most prominent conceptual frameworks

for competitive anxiety (Janelle et al., 2020). According to their con-

ceptualization, anxiety in sports should be divided into cognitive and

somatic components. Furthermore, the Sport Anxiety Scale (SAS; Smith

et al., 1990), a sport-specific multidimensional trait anxiety measure,

includes three components: somatic anxiety, worry and concentration

disruption, the latter being two facets of cognitive anxiety. The SAS re-

lies on the cognitive-affective model and demonstrates strong reliability

and validity as a measure of both cognitive and somatic anxiety related

to sport performance. Since its initial development it underwent a revi-

sion and has been employed across various sport settings (SAS-2; Smith

et al., 2006). Finally, using a multidimensional approach, the Competi-

tive State Anxiety Inventory-2 (CSAI-2; Cox et al., 2003;Martens et al.,

1990) is a widely acclaimed measure that consists of three subscales:

somatic anxiety, cognitive anxiety, and self-confidence, demonstrating

associations with athletic outcomes across diverse disciplines (e.g., Cox

et al., 2003). A recent development describes anxiety as a multidimen-

sional hierarchical construct where the general construct of competitive

anxiety includes three lower order components: cognitive (worry, pri-

vate self-focus, public self-focus), physiological (somatic tension, auto-

nomic hyperactivity), and regulatory (perceived control), reflecting the

multiple pathways described earlier (Jones et al., 2019).

Although there is a considerable number of measures capturing the

construct of anxiety in sports, as well as its associations with various

sport-related outcomes, our understanding regarding the extent to

which these findings generalize to the realm of extreme, high-risk

sports remains limited. There are several potential mechanisms that

could explain why the anxiety-sport relevant outcomes relationship

might not directly generalize to high-risk sports. For example, to consis-

tently practice such sports athletes having either very low baseline anx-

iety levels or having the necessary skills to manage the anxiety levels

might have self-selected themselves into such disciplines. Similarly, a

review highlighted that individuals with higher levels of anxiety sensi-

tivity were found to engage in less physical activity, potentially leading

them to specifically avoid these types of sports (DeWolfe et al., 2023).

Consequently, employing the anxiety measures devised for sports that

do not qualify as high-risk disciplines might fail to capture the relation-

ship between anxiety symptoms and sport-relevant outcomes in such

disciplines. In the realm of high-risk sports, tailoring the measures to re-

flect the specific demands of the respective athletic discipline is pivotal

to enable the understanding of how the anxiety - sport-relevant out-

comes relationship unfolds. A similar perspective has been emphasized

by Janelle et al. (2020) highlighting the importance of utilising sport-

specific assessment tools (Teixeira et al., 2022), allowing for a more

precise evaluation of an athlete’s mental state within the context of

their sport. Such an approach might be required to understand how

anxiety symptoms relate to various aspects of athletic practice and per-

formance in high-risk sports such as climbing or to devise targeted in-

terventions aimed at addressing the distinctive challenges faced by ath-

letes in such sports (Garrido-Palomino, I., & España-Romero, 2023). For

instance, the unique nature of climbing can be captured through ques-

tions that highlight climbing-specific fears, e.g., i2 ‘I overgrip the

holds’; i8 ‘I have mental images about falling or getting injured’; i9 ‘I

constantly try to block thoughts about falling or getting injured’; i20 ‘I

quit climbing a route because I felt terror or fear about falling’; i27 ‘I

struggle to understand what others (e.g. coach, belayer, friends) are

saying to me because the fear is too intense’; i37 ‘I feel ashamed espe-

cially when others are around (e.g. people I know, ‘better’climbers,

etc.)’.

While the existing measures provide sport psychologists and coun-

sellors with various options for capturing the different facets of sport-

related anxiety symptoms, several aspects might limit their relevance

and utility for the specific context of high-risk sports. First, such scales

mostly reflect general anxiety processes, while potentially failing to

capture sport-specific elements that might come into play within vari-

ous sport disciplines, in general, and for extreme sports, in particular.

For example, there are sports that inherently include a high degree of

risk-taking, where practitioners are routinely exposed to a higher de-

gree of objective risks, such as motorcycling, mountaineering or rock-

climbing. Although typical anxiety measures are not meant to accu-

rately capture the way in which anxiety interferes with athletic out-

comes, they lack the sensitivity needed to capture the extent of experi-

encing symptoms of anxiety in the specific context of such challenging

and highly dangerous sports. Second, extant research from other do-

mains, such as self-efficacy (Llewellyn et al., 2008) illustrates that em-

ploying contextualized, sport-specific measures of self-efficacy results

in an optimized criterion-related validity for the respective measures.

We argue that particularly for high-risk sports, the design of sport-

specific, contextualized anxiety measures is a necessity.

Consequently, we address the above-mentioned gaps by (1) devel-

oping a sport-specific anxiety measure and (2) pitting its criterion-

validity against existing sport anxiety measures, in order to illustrate

the necessity of specificity in conceptualizing and measuring anxiety

symptoms in sports.

3. Is there a need for sport specific anxiety measures?

Currently, there is limited understanding of outdoor sports that ex-

tend beyond traditional competitive formats. Examples include ‘high-

risk’sports, which have seen a significant surge in interest and partici-

pation over the past couple of decades (Clough et al., 2016). Rock-

climbing, a ‘high-risk’sport, has only been recently included in the

2021 Olympic Games, presenting a unique case as it includes both com-

petitive (indoor sport climbing, bouldering and speed), and non-

competitive formats (outdoor sport climbing, bouldering, multi-pitch,

etc.), with a majority of practitioners preferring the outdoor recre-

ational/non-competitive format (Brymer et al., 2020), as is the case in

the current study.

While in competitive formats, the actual risks faced by the practi-

tioners are greatly reduced, the practice of outdoor, non-competitive

format still poses a considerable level of risk. Consequently, the practi-

tioners of such sports are perceived as being ‘trained’to manage levels

of anxiety deemed unacceptably high in other sport contexts (e.g.,

Hunt, 1995). Previous research on anxiety indicates that elite rock-

climbers generally exhibit lower levels of anxiety symptoms than the

average population, though anxiety symptoms are still experienced

when practicing the activity (Robinson, 1985). Several investigations

showed the presence of anxiety symptoms when climbing for the less

seasoned practitioners, with anxiety symptoms emerging at various lev-

els of analysis: psychological, behavioural, and physiological assess-

ments, e.g., heart rate, difference in muscle tension/fatigue (blood lac-

tate concentration and electromyographic –EMG –activity of the mus-

cles), or ‘geometric index of entropy’(Aras & Akalan, 2014;Pijpers et

al., 2003;Pijpers et al., 2006). These results were suggestive especially

for novice climbers reporting significantly more subjective anxiety

(‘cognitive and somatic anxiety thermometer’) and objective anxiety

(exhibiting significantly higher heart rates, more muscle fatigue, and

higher blood lactate concentrations) symptoms on routes high on the

wall than when they traversed an identical route low on the climbing

wall (Pijpers et al., 2003). Furthermore, lead climbing (a practice where

climbers need to place the rope through various anchors along the route

to progress) was perceived to cause greater anxiety symptoms for the

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

athletes, compared with top-roping (where the rope is already placed

through the various anchors; Aras & Akalan, 2014;Hodgson et al.,

2009). Also, in a situation where climbers had current knowledge of the

route, lead climbing has been observed to stimulate a higher concentra-

tion of cortisol response compared to top-roping (Draper et al., 2012).

This distinction is essential for climbers because they might consider a

route completed only when done in a lead climbing format. As shown

earlier, lead climbing demands a higher level of expertise, but it also en-

tails greater danger and, consequently, increased anxiety levels.

Designing a domain specific anxiety measure for climbing will aid

researchers, coaches, and psychological professionals to develop an in-

depth understanding on whether and how anxiety symptoms interfere

with various aspects of climbing performance and, consequently, to

construct or optimize their interventions and potentially opening a new

research avenue for the investigation of anxiety in high-risk sports.

More specifically, we add to the literature on anxiety in sports in two

ways: (1) by identifying whether the construct domain of anxiety in

sports generalizes to rock-climbing, a high-risk sport and (2) by investi-

gating whether a sport-specific anxiety measure captures more con-

struct-relevant variance over and beyond the general measures of anxi-

ety in sport.

4. Aims of current research

The overarching goal of our investigation was twofold: (1) to de-

velop and validate a theoretically-anchored sport climbing inventory/

questionnaire and (2) to provide preliminary evidence on its construct

and criterion-related validity. We contribute to the general domain of

sport and athletic research by developing a sport-specific anxiety mea-

sure, focusing on uncovering whether and how anxiety symptoms come

into play in rock-climbing, a high-risk sport. From a practical stand-

point, we offer a newly developed anxiety measure that can be used ef-

ficiently when working with rock climbers. We tackled these objectives

via two different phases. Phase 1 describes the development of the

Climbing Anxiety Scale (CAS) and contains the item development and

factorial validity. Phase 2 offers preliminary evidence pertaining to the

measure’s construct, convergent, discriminatory and criterion-related

validity.

5. Phase 1: scale development and content validity

The goal of the first study was twofold: (1) to generate a sufficiently

large item pool in order to capture the most relevant aspects of climbing

anxiety (see description below), specifically somatic anxiety, cognitive

anxiety (including concentration disruption, following the multidimen-

sional anxiety theory based on the cognitive-affective model by

Martens et al., 1990), and the addition of the self-confidence subscale;

and (2) to select the most relevant items based on a content analysis

conducted by experts.

We followed Waltz and Bausell’s (1983) methodology to conduct an

item-level content analysis. The Content Validity Index (CVI) to assess

the content of individual items. We employed a panel of experts that

evaluated each item for its relevance (1 = not relevant; 4 = very rele-

vant), clarity (1 = not clear; 4 = very clear), simplicity (1 = not simple;

4 = very simple), and ambiguity (1 = doubtful; 4 = meaning is clear) us-

ing a 4-point scale. The proportion of experts who rate an item was cal-

culated with CVI scores of < 0.75 generally considered strong.

6. Method

6.1. Participants

Phase 1 of the study involved the participation of four female sport

psychologists, one female clinical psychologist, and one male coach, all

of whom were specialized in working with climbers, and were aged be-

tween 29 and 46 years old. The sport psychologists’experience ranged

between 3 and 7 years, the clinical psychologist had 20 years’experi-

ence, and the coach had 10 years’coaching experience. A total of six ex-

perts from the United Kingdom (n= 4), Austria (n= 1), and Slovenia

(n= 1) participated in the study. All the experts were fluent in English

at a C2 level and were independent of the research team.

6.2. Procedure

The study was conducted in accordance with the Declaration of

Helsinki and approved by the Ethics Committee of [MASKED FOR RE-

VIEW] University. Participants were recruited from the International

Association of [MASKED FOR REVIEW], they volunteered to respond to

our invitation via an online survey and were required to provide writ-

ten informed consent before participating in the study. No compensa-

tion was offered.

6.3. Climbing anxiety scale (CAS)

6.3.1. Model development and initial item pool

Item development. The principal investigator, an experienced clinical

psychologist and rock-climbing practitioner (over 20 years of sport

practice) together with a psychometrician having extensive expertise

in test development and adaptation (the corresponding author) formu-

lated an initial item pool of 45 items. To facilitate further reviews and

data collection, all the items were written in English. In developing the

items, an extensive review of the theoretical and empirical literature

on anxiety in sports and athletic performance was conducted. The

items were subsequently formulated by following a rational-theoretical

approach, being geared towards capturing the facets of sport-related

anxiety (such as the multidimensional anxiety theory with cognitive or

somatic aspects; Martens et al., 1990) as well as components of the

cognitive affective model (Smith et al., 2006). The CAS was devised as

an anxiety trait measure specific for rock-climbing. The participants re-

ceived the following set of instructions: ‘Many climbers get tense or ner-

vous before or during their climb. This happens even to pro climbers. Please

read each statement, then tick the appropriate box that best represents how

you usually feel during climbing. There are no right or wrong answers. Do

not spend too much time on any single statement but select the option that

most accurately depicts your general feelings during climbing’. Moreover,

the total number of items included in the initial item pool reflected a

balance between parsimony (measuring anxiety symptoms via a rela-

tively short and easy to administer inventory) and content relevance

(employing a sufficiently large number of items to reflect the potential

facets of anxiety as manifested in rock-climbing). No strict limitation

was imposed on the total number of items. The items captured the fol-

lowing constructs: somatic anxiety (16 items), cognitive anxiety (20

items), concentration disruption (4 items), and self-confidence (5

items). Out of the 45 items developed, we used and reformulated eight

items (i) form CSAI-2R (see Appendix 1 CAS-45: i10, i11, i14, i16, i26,

i30, i36, i41) and five items from SAS-2 (see Appendix 1 CAS-45: i17,

i19, i28, i29, i42). The eight items based on the CSAI-2R state anxiety

measure, were modified in order to reflect trait anxiety levels in climb-

ing situations, for example, the item ‘I’m concerned about performing

poorly’(i11 from CSAI-2R) was modified into a trait item by adding

the indication ‘how you usually feel during climbing’to ‘I’m concerned

about climbing poorly’(i11 from CAS-45). The remaining 32 items

were formulated to capture somatic anxiety (i2, i4, i8, i24, i31, i33,

i34, i39, i40, i43, i44, i45), cognitive anxiety (i1, i3, i5, i6, i7, i9, i12,

i13, i15, i18, i20, i21, i22, i25, i27, i32, i35, i37, i38), and self-

confidence (i23). The following principles were followed during the

item generation process: (a) Content coverage –the items were formu-

lated to reflect potentially unique construct-relevant aspects and be-

haviours; (b) Parsimony –anxiety in sport in general and in sport

climbing more specifically, is not described in a literature as being a

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

broad construct that includes a hierarchical latent structure or a com-

plex measurement architecture. Consequently, the construct's relative

parsimony should be reflected in the scale's structure and number of

items; (c) Spearman-Brown principle was followed in formulating the

number of items, thus for each potential subdimension/facet we estab-

lished a minimum five number of items; (d) Scale length –we focused

on obtaining the minimal number of items that could reflect via unique

contents the various facets of climbing anxiety (cognitive, emotional

and somatic). The final item set included a total of 45 items which

were analysed by a panel of six experts.

A relatively standard set of scale completion instruction was used:

‘Many climbers get tense or nervous before or during their climb. This

happens even to pro climbers. Please read each statement, then tick the

appropriate box that best represents how you usually feel during climb-

ing. There are no right or wrong answers. Do not spend too much time

on any single statement but select the option that most accurately de-

picts your general feelings during climbing’.

6.4. Data analysis

To assess content validity, the experts provided ratings for the items

using a 4-point CVI scale (Waltz & Bausell, 1983). The criteria can be

found in Appendix 1. Each expert rated each item according to the crite-

ria. To calculate the CVI, the number of responses for each item rated as

three or four was summed and divided by the total number of items, re-

sulting in a fractional proportion. Items with a CVI of 0.75 or higher

were contemplated to have adequate content validity.

7. Results

Appendix 1 displays the Mean CVI scores for each item for rele-

vance, clarity, simplicity, and ambiguity. In total, six items had mean

CVIs below 0.75 (items 4, 6, 8, 10, 24, 26) and were consequently re-

moved from the item pool. Each of these six items were reviewed to

determine if they could be revised, but they could not have been re-

vised without replicating another item and therefore they were re-

moved, yielding a CVI scores for all non-revised 39 items ranged from

0.83 to 1.

8. Phase 2: scale validation

The aim of the second phase was to examine the internal structure of

the scale generated in Phase 1. We tested the preliminary factor struc-

ture of the newly developed measure and refined it through an iterative

process. The overarching aim of this phase was to derive a new and psy-

chometrically coherent assessment of climbing anxiety. This aim was

further split into three separate objectives: (1) refining the initial item

pool by analysing the response distributions per item; (2) estimating the

measure’s reliability by examining its internal consistency and the

item-total correlations and (3) examining three different forms of valid-

ity: construct validity by deploying a series of exploratory and confir-

matory factorial analyses; convergent and discriminant validity by ex-

amining the scale’s associations with sport anxiety, anxiety and depres-

sion, and climbing self-efficacy; and criterion-related validity by exam-

ining its relationship with rock-climbing performance.

9. Method

9.1. Participants

Power estimation. To determine the necessary sample sizes, we em-

ployed R software, version 4.1.2 (R Core Team, 2021)‘semPower’

package, relying on the approach outlined by Jobst et al. (2023). The

power analysis was based on the following assumptions: 20 observed

variables, df = 167, RMSEA = 0.05, loadings of 0.80 and suggested

that the minimum sample size to obtain a power of 0.80 is N = 123.

Following institutional ethics approval [MASKED FOR REVIEW],

we collected data from 153 participants, regular practitioners of rock-

climbing (including sport climbing and/or bouldering as two different

forms of rock-climbing practice). The sample included 95 males

(62.1%) and 58 females (37.9%), with ages between 17 and 67

(M= 33.18, SD = 10.75). At this stage, the single inclusion criterion

was a minimum of 12 months experience in practicing rock-climbing.

Their experience in rock-climbing ranged between 2 and 42 years

(M= 11.18, SD = 9.32). In respect to education, 79 participants

(51.63%) attained a graduate or post-graduate degree, 44 participants

(28.75%) attained a bachelor’s degree, and the remaining 30 partici-

pants (19.61%) had a high school degree or equivalent. Participants

were polled from 18 different countries from Asia, Australia, Europe,

North America, and South America.

9.2. Measures

Depression Anxiety Stress Scales Short Form (DASS-21; Lovibond &

Lovibond, 1995). DASS-21 is a brief version of DASS-30 which is a self-

report measure of negative emotional states of depression (e.g., ‘I felt

that life was meaningless’), anxiety (e.g., ‘I was worried about situa-

tions in which I might panic and make a fool of myself’), and stress

(e.g., ‘I found it hard to wind down’), measured on a 4-point scale from

0 (did not apply to me at all) to 3 (applied to me very much or most of the

time). Higher scores indicate higher level of depression, anxiety, and

stress, respectively. Internal consistencies for the current sample

ranged between 0.92 (depression), 0.74 (anxiety), and 0.82 (stress).

Sport Anxiety Scale-2 (SAS-2; Smith et al., 2006). Sport anxiety was

assessed by employing the 15-item inventory that measures the compet-

itive trait anxiety which is usually experienced by athletes before or

during competitions. To stay consistent with our sport and because par-

ticipants were not competing, we used the term ‘climb’instead of ‘play’

or ‘game’. SAS-2 is composed by 3 subscales: somatic anxiety (e.g., ‘My

muscles feel tight because I am nervous’), worry (e.g., ‘I worry that I

will climb badly’), and concentration disruption (e.g., ‘It is hard to con-

centrate on the climb’), measured on a 4-point scale from 1 (not at all)

to 4 (very much). Higher scores indicate higher level of somatic anxiety,

worry, and concentration disruption, respectively. Internal consisten-

cies for the current sample ranged between 0.83 (somatic anxiety), 0.88

(worry), and 0.88 (concentration disruption).

Climbing Self-Efficacy Scale (CSES; Llewellyn et al., 2008). The CSES

is a scale that has been previously validated for the climbing domain. It

evaluates self-efficacy in relation to the various sub-skills needed for

successful performance in climbing. Participants rate how confident

they feel, in general, from 0% (not at all confident) to 100% (extremely

confident) regarding 10 non-hierarchical themes (e.g., ‘Prepare physi-

cally for demanding routes’;‘Prepare mentally for demanding routes’;

‘Use appropriate climbing techniques’). The total score ranges from 0

to 1000, where higher scores indicate greater confidence that is higher

level of self-efficacy in climbing. For CSES internal consistency for the

current sample was 0.93 (Cronbach’s alpha).

Climbing Anxiety Scale (CAS-39). Climbing anxiety was assessed by

using the 39-item scale developed in Phase 1. On a 4-point scale from 1

(not at all) to 4 (very much so), participants rate how they feel in gen-

eral before or during climbing. CAS is composed by 3 dimensions: cog-

nitive anxiety (e.g., ‘I find it hard to concentrate on the climb because

of my inner fears’), somatic anxiety (e.g., ‘I feel jittery’), and self-

confidence (‘I feel confident while climbing’). Higher scores indicate

higher level of somatic anxiety, cognitive anxiety, and self-confidence,

respectively.

Sport Climbing Performance. In estimating rock-climbing perfor-

mance, two components are taken into account: route difficulty level

and climbing style. In respect to estimating difficulty, different scales

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

have been proposed by various national or international mountaineer-

ing and/or climbing associations: the French/sport scale, the Yosemite

Decimal System (YDS), the Union Internationale des Associations

d’Alpinisme scale (UIAA), and the Ewbank scale, following the ap-

proach used in ([Masked, 2022]). Drawing from an international sam-

ple of rock-climbing participants, each participant was allowed to se-

lect the most familiar reporting scale (French, UIAA, YDS or Ewbank).

The reported grades have been converted according to the recommen-

dations outlined in IRCRA’s position statement (IRCRA; Draper et al.,

2015). We asked each participant to report the most difficult routes

that they managed to successfully complete in the three different styles

outlined above during the past 12 months.

9.3. Procedure

The measures were administered via an online survey. Considering

the challenges in identifying and enrolling a sufficiently large number

of practitioners from a single country, we decided to open the survey to

the international community of rock climbing and bouldering practi-

tioners. Consequently, the survey was disseminated via email and social

networks to English-speaking adult climbers from different countries.

The participants were explained their rights as volunteers and all of

them provided their informed consent before completing the survey.

Fifteen randomly allocated prizes were offered. The prizes included: 4

books in the field of climbing and 11 sessions for optimizing the psycho-

logical aspects related to climbing performance, facilitated by the main

author ([Masked]).

9.4. Data analysis

Preliminary analyses were used in order to refine the initial item

pool and to gain insights in respect to the measure’s structure and valid-

ity. We further refined the item pool by reviewing the response distribu-

tions and estimating the measure’s internal consistency. Subsequently,

to examine the factorial structure of the newly developed measure, we

proceeded as follows: first, we analysed the Kaiser–Meyer–Olkin and

Bartlett’s test of sphericity, then we established the number of factors

by conducting a parallel analysis (Horn, 1965) and visually inspecting

the screeplot. Next, an exploratory factorial analysis (EFA) with a maxi-

mum likelihood extraction with an oblimin rotation was deployed. We

further refined the item pool by eliminating items with very small load-

ings, items that cross-loaded onto non-target factors or having a low

communality (Child, 2006). Then we ran an additional EFA on the re-

maining item set. Although the general guidelines suggest running the

confirmatory factorial analysis (CFA) on a different sample, we opted to

run the CFA to offer preliminary information on the measure’s factorial

validity, as a confirmation sample drawn from such a particular popula-

tion was difficult to obtain in a reasonable timespan. To inform about

the measure’s convergent and divergent validity we analysed the mean

squared correlations between the latent factors as well as the average

variance explained. Hierarchical regressions were employed to estimate

criterion-related validities.

We then conducted an Item Response Theory (IRT) analysis to look

into the quality of the items and the test in its entirety; we computed

discrimination and difficulty parameters under the auspices of the poly-

tomous Generalized Partial Credit Model (GPCM) (Muraki, 1992), and

inspected various item information and test information indicators.

Finally, we conducted a network analysis for a supplementary indi-

cator of item quality (specifically, centrality, betweenness, closeness,

strength, and expected influence), and of item grouping (a visual indi-

cator of relationships between items for construct validity).

All analyses were performed with IBM-SPSS version 26 (IBM Corp.,

2019) and Rsoftware, version 4.1.2 (R Core Team, 2021), using the

packages ‘psych’,‘nfactors’,‘factoextra’ ‘lavaan’,‘semPower’,‘mirt’,

‘qgraph’and ‘bootnet’.

10. Results

Means, standard deviations, bivariate correlations between our

main variables and internal consistency reliabilities (Cronbach’s alpha)

are shown in Table 1 and discussed below. The relatively high correla-

tions between the three anxiety subscales suggest that our analyses

might have been affected by multicollinearity. Consequently, we deter-

mined the Variance Inflation Factors (VIF) for each of the three sub-

scales. The VIFs for each subscale were below the 2.5 threshold

(Johnston et al., 2018); for CAS Factor 1 (cognitive anxiety),

VIF = 1.29, for CAS Factor 2 (somatic anxiety), VIF = 1.30, and for

CAS Factor 2 (self-confidence), VIF = 1.23.

10.1. Initial validation

Step 1. Response distribution. In this step we removed the items that

had a range of less than 4 steps, e.g., where the answers were

located between 1 and 3, respectively between 2 and 4 or 3 and 5.

Following this we removed item 33 (Appendix 1) because it had a

very narrow range (min = 1; max = 3).

Step 2. Internal reliability analysis. After removing the aforementioned

item, the internal consistency reliability was computed (Cronbach’s

alpha = 0.87). We removed ten items with item-total correlations

less than 0.30: (Appendix 1): i1 (r= 0.27), i7 (r = 0.27), i11

(r= 0.11), i13 (r= 0.17), i17 (r= 0.06), i22 (r= 0.16), i28

(r= 0.29), i29 (r= 0.19), i31 (r= 0.21), i32 (r= 0.23), i33

(r= 0.26).

To estimate reliability for the refined version of the Climbing Anxi-

ety Scale in a more robust manner McDonald's omega was computed.

The omega coefficient for the overall CAS measure was found to be

ω= 0.94, suggesting a good reliability (McDonald, 1999). The omega

coefficients for the three factors were 0.85 (Factor 1), 0.88 (Factor 2)

and 0.77 (Factor 3). The omega coefficients for each factor were re-

ported in Appendix 2. The item-factor loadings corresponding to the re-

liability analysis were reported in Appendix 2 (Schmid Leiman Factor

loadings matrix).

Step 3. Exploratory Factor Analysis (EFA). To examine the factorial

structure of the newly developed measure, we analysed the latent

factor structure by deploying a series of EFAs on the dataset.

Next, we examined three different forms of validity, and

proceeded as follows:

(a) Factor Structure and Factorial validity. Considering that the

measure was developed on a theoretical measurement

foundation, we expected the data to suggest a three-factor

component (e.g., Cox et al., 2003). The Kaiser–Meyer–Olkin

coefficient yielded a value of 0.91, and Bartlett’s test of

sphericity was significant (p < 0.001), indicating that the

dataset has an underlying factorial structure (Tabachnick &

Fidell, 2007). Subsequently a maximum likelihood extraction

with an oblimin rotation was conducted. The three-factor solution

accounted for 43% of the total variance, the first factor

explaining 22%, the second only 8% and the third 13%. After

analysing the pattern of item loadings reported in Table 2, we

eliminated the items having a low communality (<0.20) and/or

a pattern of cross-loadings amongst the three different factors

(Child, 2006). An item was considered to cross-load if it loaded

0.32 or greater on one factor and the difference between loadings

on other factors was <0.10 (Brown, 2015). After applying the

criteria outlined above, we removed the following items due to

low communality (<0.20): i5, i21, i25, i26, i39. Additionally, we

removed i10, i14, i15 due to cross-loading (≥0.32, or difference

between loadings on other factors <0.10). Altogether we

removed eight items: i5, i10, i14, i15, i21, i25, i26, i39 (see

CORRECTED PROOF

Table 1

Descriptive statistics, scale intercorrelations and internal consistency.

M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 Age 32.1 10.0 –

2 Experience 9.0 9.1 0.69*** –

Anxiety measures

3 CAS Overall 16.0 4.5 0.15* 0.46*** (0.92)

4 CAS Factor 1 17.0 4.6 0.19** 0.48*** 0.99*** (0.91)

5 CAS Factor 2 17.3 5.0 0.18** 0.48*** 0.98*** 0.90* (0.80)

6 CAS Factor 3 16.5 4.3 0.08 0.35*** 0.94*** 0.73* 0.59* (0.76)

7 DASS-21 Anxiety 15.0 3.7 0.13 0.45*** 0.96*** 0.93*** 0.92*** 0.85*** (0.74)

8 DASS-21 Stress 10.9 9.9 −0.03 0.14* 0.34*** 0.43*** 0.34*** 0.35*** 0.35*** (0.82)

9 DASS-21 Depression 19.5 4.1 0.26** 0.48*** 0.59*** 0.68*** 0.59*** 0.61*** 0.66*** 0.98*** (0.92)

10 SAS-2 Overall 11.5 10.1 −0.03 0.14* 0.35*** 0.44*** 0.35*** 0.35*** 0.36*** 0.99*** 0.94*** (0.90)

11 SAS-2 Somatic 19.1 4.2 0.18* 0.33*** 0.42*** 0.48*** 0.42*** 0.48*** 0.50*** 0.85*** 0.77*** 0.72*** (0.83)

12 SAS-2 Worry 18.1 3.7 0.21** 0.44*** 0.58*** 0.63*** 0.53*** 0.60*** 0.62*** 0.96*** 0.95*** 0.90*** 0.75*** (0.88)

13 SAS-2 Concentration Disruption 15.5 5.1 −0.13* −0.13* −0.06 −0.02 −0.05 −0.07 0.01 −0.04 −0.17* −0.03 −0.29*** −0.17* (0.88)

14 CSES 66.60 16.39 −10 0.18* −0.39** −0.33** −0.22** −0.22** −0.10 −0.09 −0.14 −0.23** −0.29** −0.21** −0.22** (0.93)

15 Climbing Performance Overall 3.4 0.6 0.23** 0.20** 0.21** 0.24** 0.22** 0.21** 0.17* 0.09 0.22** 0.08 0.17* 0.24* −0.30*** 0.36** (0.79)

Note: N = 153. *p< 0.05; **p< 0.01; ***p< 0.001. CAS = Climbing Anxiety Scale; DASS-21 = Depression Anxiety Stress Scales 21; SAS-2 = Sport Anxiety Scale 2; CSES = Climbing Self-Efficacy Scale.

Internal consistency estimates (α) are shown in parentheses along the diagonal.

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

Table 2

Item factor loadings and item communalities.

First round EFA Second round EFA

Factor

h2Factor

Factor

Item 2 −0.11 0.64 0.12 0.38 −0.05 0.54 0.11 0.31

Item 3 0.64 0.09 0.15 0.62 0.62 0.12 0.17 0.64

Item 4 0.65 0.11 0.16 0.67 0.63 0.12 0.19 0.69

Item 5 −0.06 0.45 0.13 0.20 – – – –

Item 6 0.67 0.05 0.08 0.56 0.45 0.28 0.14 0.56

Item 8 0.57 −0.01 0.03 0.34 0.67 −0.07 0.03 0.40

Item 9 0.72 −0.09 −0.03 0.42 0.75 −0.09 −0.02 0.47

Item 10 0.41 0.30 −0.03 0.42 – – – –

Item 12 0.25 −0.05 0.56 0.48 0.09 0.07 0.61 0.49

Item 14 0.37 0.31 0.01 0.40 – – – –

Item 15 0.39 0.29 0.20 0.56 – – – –

Item 16 0.77 0.03 0.02 0.66 0.53 0.27 0.10 0.63

Item 18 0.89 −0.04 0.04 0.79 0.63 0.21 0.13 0.74

Item 19 0.08 −0.01 0.47 0.26 0.02 0.00 0.51 0.27

Item 20 0.74 −0.01 −0.12 0.47 0.80 −0.05 −0.10 0.80

Item 21 0.14 0.30 0.02 0.18 – – – –

Item 23 0.22 0.59 −0.07 0.54 0.11 0.69 −0.06 0.55

Item 24 0.18 0.50 0.08 0.46 −0.07 0.73 0.12 0.54

Item 25 0.28 0.13 0.08 0.19 – – – –

Item 26 −0.05 0.41 0.12 0.18 – – – –

Item 27 0.44 0.23 −0.16 0.32 0.28 0.40 −0.14 0.33

Item 30 0.14 0.01 0.58 0.45 0.16 −0.06 0.61 0.45

Item 34 0.47 0.27 0.01 0.49 0.29 0.45 0.05 0.50

Item 35 −0.04 0.03 0.84 0.69 −0.04 −0.02 0.83 0.66

Item 36 0.04 0.67 −0.09 0.46 0.04 0.65 −0.10 0.41

Item 37 0.02 0.44 0.15 0.28 −0.02 0.47 0.15 0.28

Item 38 0.12 0.40 −0.06 0.22 0.03 0.48 −0.06 0.23

Item 39 −0.02 0.39 −0.01 0.14 – – – –

Factor 1 – – – – – – – –

Factor 2 0.71 – – – 0.70 – – –

Factor 3 0.49 0.34 – – 0.49 0.42 – –

Variance

22% 8% 13% –21% 10% 17% –

Note: EFA = Exploratory Factor Analysis.

Table 2). Next, we examined the item factor loadings and the

item parameters. The results were included in Table 2, Item

Factor Loadings and Item Communalities. A second exploratory

analysis on the remaining 20-item data set by employing the

previously outlined approach. The solution accounted for a total

of 48% of variance and no further cross-loadings and low

communalities were identified.

Next, we conducted CFA based on a maximum likelihood estima-

tion. Because the χ2-test is sensitive to sample sizes, the root mean

square error of approximation (RMSEA), standardized root-mean-

square residual (SRMR) and the comparative fit index (CFI) were used

to estimate goodness-of-fit (Browne & Cudeck, 1992). In analysing

goodness of fit the thresholds of 0.90 or higher for CFI and 0.08 or

lower for RMSEA and SRMR were employed (Browne & Cudeck, 1992;

Hu & Bentler, 1999). The three-factor solution provided a good fit for

the data, CFI = 0.92, RMSEA = 0.06, 90% C.I. [0.05 - 0.08], and

SRMR = 0.06. Table 3 includes the final item list with their corre-

sponding factor loadings.

With respect to the measure’s latent factor structure, three compo-

nents were identified as: Factor 1 –Cognitive Anxiety (e.g., i3: ‘I find it

hard to concentrate on the climb because of my inner fears’); Factor 2 –

Somatic Anxiety (e.g., i23 ‘I feel tense or rigid in my movements’); and

Factor 3 –Self-confidence (e.g., i12 “I feel confident while climbing”).

(b) Convergent and discriminant validity was examined by analysing

the scale’s associations with latent factors for anxiety levels,

depression and stress, sport anxiety, and climbing self-efficacy as

Table 3

Loadings matrix.

Item Factor

Factor

Item

I overgrip the holds. –1–

Item

I find it hard to concentrate on the climb because

of my inner fears.

1– –

Item

I struggle to stay focused on the climb because of

intense fears.

0.90 – –

Item

I struggle to maintain my composure. 0.86 – –

Item

I have mental images about falling or getting

injured.

0.76 – –

Item

I constantly try to block thoughts about falling or

getting injured.

0.82 – –

Item

I feel confident while climbing. ––1

Item

I feel agitated or fearful. 0.92 – –

Item

I struggle to keep calm. 0.90 – –

Item

I feel confident in my ability to handle the

unexpected challenges.

––0.75

Item

I quit climbing a route because I felt terror or fear

about falling.

0.96 – –

Item

I feel tense or rigid in my movements. –1.32 –

Item

I feel jittery. –1.33 –

Item

I struggle to understand what others (e.g. coach,

belayer, friends) are saying to me because the fear

is too intense.

–0.68 –

Item

I’m confident because I visualize myself reaching

my goal.

––0.95

Item

I start shaking or trembling uncontrollably. –1.08 –

Item

I’m confident about performing well while

climbing.

––1.09

Item

My muscles are tensed. –1.15 –

Item

I feel ashamed especially when others are around

(e.g. people I know, ‘better’climbers, etc.)

–1.13 –

Item

My palms sweat nervously. –1.01 –

presented in Appendix 3. Significant and mid-sized correlations

were observed between the newly established anxiety scale, CAS,

its three factors and the other measures (Appendix 3).

Additionally, to inform about the measure’s level of convergence

between its factors, as well as its convergence with the other

measures, the Average Variance Extracted (AVE) was estimated.

We employed Fornell and Larcker’s (1981) criteria, AVE values

higher than 0.50 indicating that the underlying constructs explain

at least half of the observed variance. For the overall measure, the

AVE was 0.51. The average variances extracted for the three

different factors were 0.68 for Factor 1, 0.41 for Factor 2, and

0.69 for Factor 3 (see Table 4). To inform about the measure’s

discriminant validity, we contrasted the AVE with the squared

correlations between the latent factors across all the measures. As

expected, the differences between the AVE for each factor and the

squared correlations with the SAS-2 were smaller compared to

the ones observed for DASS-21 and CSES. While the AVE to

r−2differences between Factors 1 and 3 and SAS-2 were higher,

suggesting that these two components might have discriminant

validity in respect to the general construct of sport anxiety, for

Factor 2 the r1were higher than the AVE, indicating that Factor 2

partly overlaps with constructs captured by the SAS-2. More

precisely, Factor 2 overlapped to a higher degree with somatic

anxiety. With respect to the remaining measures, as expected, the

r−2were generally lower than the AVEs for each dimension of the

newly developed measure.

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

Table 4

Convergent and discriminant validity –climbing anxiety scale.

Measure Scales Factor 1 Factor 2 Factor 3

CAS Factor 1 –

Factor 2 0.82 –

Factor 3 0.53 0.34 –

SAS-2 Somatic 0.52 0.61 0.28

Worry 0.17 0.28 0.19

Concentration Disruption 0.49 0.58 0.30

r−20.39 0.49 0.25

DASS-21 Depression 0.06 0.08 0.04

Anxiety 0.13 0.16 0.13

Stress 0.13 0.15 0.11

r−20.11 0.13 0.09

CSES Climbing Self-Efficacy 0.14 0.11 0.23

r−2across measures 0.23 0.28 0.18

AVE 20.68 0.41 0.69

Note:r−2= mean squared correlations between latent factors; AVE = Average

Variance Extracted. CAS = Climbing Anxiety Scale; SAS-2 = Sport Anxiety

Scale 2; DASS-21 = Depression Anxiety Stress Scales 21; CSES = Climbing

Self-Efficacy Scale.

where the overall climbing anxiety score was employed as a

predictor of overall climbing performance over and beyond age,

gender, and the other variables. Next, we transformed the raw

scores in latent factor scores. We employed hierarchic regression,

following recommendations in the literature (Hunsley & Meyer,

2003). Age, gender and experience were entered in step 1. In step

2 we introduced climbing self-efficacy, the CSES. In step 3 we

introduced sport anxiety, the SAS-2. Climbing anxiety, the CAS

was added in step 4 in order to estimate its validity over

demographics, climbing self-efficacy, and sport anxiety. The

results were reported in Table 5. Four dimensions of climbing

performance were considered: overall climbing performance

(latent factors derived via CFA), highest redpoint, highest

onsight, and highest flash.

Outdoor Sport Climbing Performance. In step 2, climbing self-efficacy

(β= 0.30, p< 0.001) significantly predicted overall sport climbing

performance (latent) and accounted for 7.9% of the overall variance in

overall sport climbing performance over age, gender, and experience

(ΔR2= 0.079, p< 0.001). Climbing self-efficacy (β= 0.34,

p< 0.001) significantly predicted highest redpoint and accounted for

10.5% of the overall variance over age, gender, and experience

(ΔR2= 0.105, p< 0.001). Climbing self-efficacy (β= 0.30,

p< 0.001) significantly predicted highest onsight and accounted for

8.2% of the overall variance over age, gender, and experience

(ΔR2= 0.082, p< 0.001). Finally, climbing self-efficacy (β= 0.32,

p< 0.001) significantly predicted highest flash and accounted for

9.1% of the overall variance over age, gender, and experience

(ΔR2= 0.091, p< 0.001). In step 3, when adding sport anxiety, the

models were not significant for any performance indicators. When in-

cluded in step 4, the dimensions that significantly predicted overall

sport climbing performance (latent) were climbing self-efficacy

(β= 0.21, p< 0.01), sport anxiety (β= 0.20, p< 0.05), and climb-

ing anxiety (β=−0.07, p< 0.05). This accounted for a mere 4% of

the overall variance in sport climbing performance over age, gender,

experience, climbing self-efficacy, and sport anxiety (ΔR2= 0.040,

p< 0.01). In step 4, the dimensions that significantly predicted high-

est redpoint were climbing self-efficacy (β= 0.21, p< 0.05). Further-

more, in step 4, the dimensions that significantly predicted highest on-

sight were climbing self-efficacy (β= 0.20, p< 0.05) and climbing

anxiety (β=−0.08, p< 0.05). This accounted for 4.7% of the overall

variance in highest onsight over age, gender, experience, climbing self-

efficacy, and sport anxiety (ΔR2= 0.047, p< 0.01). Finally, in step 4,

the dimensions that significantly predicted highest flash were climbing

self-efficacy (β= 0.24, p< 0.05) and climbing anxiety (β=−0.07,

p< 0.05). This accounted for 3.2% of the overall variance in highest

flash over age, gender, experience, climbing self-efficacy, and sport

anxiety (ΔR2= 0.032, p< 0.01).

10.2. Item Response Theory analysis

Extraction of the IRT model yielded good fit indices for a unidimen-

sional model: RMSEA = 0.075; TLI = 0.938, CFI = 0.946. Table 6 pre-

sents the item parameters, i.e., a (discrimination) and b1, b2, and b3

(difficulty). Overall difficulty is computed as a mean of the 3 difficulty

parameters. Deviance is the difference between the respective difficulty

Table 5

Climbing anxiety’s incremental validity over Age, Gender, Experience, Climbing Self-Efficacy Scale, and Sport Anxiety Scale in predicting Outdoor Sport Climbing

Performance.

Step Rock-climbing performance indicators

Overall Sport Climbing Performance (latent) past 12 months Highest Redpoint (past 12 months) Highest Onsight Highest Flash

Predictors βR2ΔR2βR2ΔR2βR2ΔR2βR2ΔR2

1 Age −0.40*** 0.161 – −0.43*** 0.204 – −0.49*** 0.220 – −0.43*** 0.146 –

Gender −0.19* −0.27** −0.20* −0.17*

Experience 0.43*** 0.45*** 0.54*** 0.43***

2 Age −0.38*** 0.240 0.079*** −0.40*** 0.308 0.105*** −0.46*** 0.302 0.082*** −0.39*** 0.237 0.091***

Gender −0.11 −0.18* −0.13 −0.08

Experience 0.38*** 0.38*** 0.48*** 0.36***

CSES 0.30*** 0.34*** 0.30*** 0.32***

3 Age −0.38*** 0.240 0.000 −0.40*** 0.315 0.007 −0.46*** 0.302 0.000 −0.39*** 0.238 0.001

Gender −0.11 −0.18* −0.13 −0.08

Experience 0.39*** 0.37*** 0.47*** 0.36***

CSES 0.30*** 0.33*** 0.30*** 0.32***

SAS-2 0.01 −0.09 −0.01 −0.01

4 Age −0.33** 0.280 0.040** −0.38*** 0.330 0.015 −0.41*** 0.349 0.047** −0.33*** 0.269 0.032**

Gender −0.05 −0.15 −0.09 −0.06

Experience 0.34*** 0.33** 0.42*** 0.30***

CSES 0.21** 0.21* 0.20* 0.24*

SAS-2 .20* 0.09 0.17 0.15

CAS-20 −0.07* −0.05 −0.08* −0.07*

Note:β=Standardized β; *p< 0.05; **p< 0.01; ***p< 0.001. CSES = Climbing Self-Efficacy Scale; SAS-2 = Sport Anxiety Scale 2; CAS-20 = Climbing Anxiety

Scale –20 items.

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

Table 6

Item parameters estimated by the IRT analysis (Generalized Partial Credit

Model).

Item a b1 b2 b3 Overall difficulty d.b1 d.b2 d.b3

Item 2 0.88 −0.66 1.04 3.50 1.29 −1.95 −0.25 2.20

Item 3 2.29 −0.07 1.11 1.51 0.85 −0.92 0.26 0.66

Item 4 2.94 0.43 1.41 1.88 1.24 −0.81 0.17 0.64

Item 6 1.91 −0.12 1.14 2.11 1.05 −1.16 0.10 1.07

Item 8 0.85 −0.07 1.87 1.10 0.97 −1.04 0.90 0.14

Item 9 1.02 0.47 1.41 1.54 1.14 −0.67 0.27 0.40

Item 12 0.85 −0.69 1.11 2.41 0.94 −1.63 0.17 1.47

Item 16 2.37 0.29 1.11 2.03 1.14 −0.85 −0.03 0.88

Item 18 3.94 0.13 1.25 2.18 1.19 −1.06 0.07 0.99

Item 19 0.52 −1.71 0.73 3.13 0.71 −2.43 0.01 2.41

Item 20 0.89 0.92 0.96 0.98 0.95 −0.03 0.00 0.03

Item 23 1.32 −0.16 1.50 2.62 1.32 −1.48 0.18 1.30

Item 24 1.15 0.17 1.57 2.22 1.32 −1.15 0.25 0.90

Item 27 1.29 1.84 2.41 3.34 2.53 −0.69 −0.12 0.81

Item 30 0.87 −2.21 −0.26 1.56 −0.30 −1.91 0.05 1.86

Item 34 1.74 1.04 1.98 2.36 1.79 −0.76 0.19 0.57

Item 35 0.74 −2.02 0.61 1.87 0.15 −2.17 0.46 1.71

Item 36 0.90 −0.36 1.49 3.18 1.44 −1.80 0.05 1.74

Item 37 0.69 0.74 1.53 2.11 1.46 −0.72 0.07 0.65

Item 38 0.52 1.45 2.25 1.61 1.77 −0.32 0.48 −0.16

Note: a = discrimination; b1, b2, b3 = difficulty; d.b1, d.b2, d.b3 = deviation

parameters.

parameter and the overall difficulty. An analysis of discrimination para-

meters shows very strong discrimination for some items (e.g., i3, i4,

i16, i18) and weaker, though acceptable discrimination for others (e.g.,

i37, i38 etc.). An analysis of the difficulty parameters shows that the

mean difficulty is 1.15; a single item (i30) has difficulty below 0; all the

other items have over-average difficulty of around 1. Figures 1a and 1b

also visualize the item traces and item information brought by each of

the test's items and presents in a more compelling manner the same in-

formation, showing both differences in discrimination between the var-

ious items, and the fact that most items cover the average and medium-

to-high areas with more information (roughly the area between

θ= [0.00, 2.00]). Figure 1c presents the Test Information Function for

the entire test, while Figure 1d visualizes the standard error of measure-

ment for the test score, conditional on ability level. These figures show

that the test discriminates very well at “ability levels”(i.e., symptom

levels) of θ= [−0.50, 3.00], with the lowest possible measurement er-

rors at symptom levels of θ= [0.00, 2.00]. This was already suggested

by the item information curves and other item statistics.

10.3. Network analysis

Table 7 visualizes the node (item) parameters of betweenness,

closeness, strength, and expected influence resulted from the network

analysis. Figure 1e visualizes the centrality chart for the nodes (items),

and Figure 1f visualizes the network plot showing the various relation-

ships (in form of proximity of nodes) and strengths of relationships

(i.e., the line width of lines spanning the nodes). These results confirm

in principle the structure of the test in 3 separate but closely connected

clusters of items. A more clearly separated cluster 3 (i12, i19, i30, and

i35) is visible, having i12 as the lead item, and corresponding to sub-

scale 3 (self-confidence). Then, a cluster 2, corresponding to subscale 2

(somatic anxiety), containing 2 sub-clusters, namely 2a (i2, i23, i24,

i36, i37) and 2b (i27, i34, i38). Finally, a cluster 1, corresponding to

subscale 1 (cognitive anxiety), containing the remaining items, some of

these seem to also be the most influential items of the entire test (e.g.,

i3, i4, i18), both in terms of relative influence (e.g., relative influence

[i18] = 2.30) and of strength (e.g., strength[i18] = 2.32), and in

terms of relationships to other items (e.g., betweenness[i3] = 2.81;

closeness[i3] = 1.78).

Figure 1. (a) Item traces for the test’s items. (b) Item information function for the test’s items. (c) Test Information Function. (d) Standard error of measurement for

the test score, conditional on ability level. (e) Centrality chart for the nodes (items). (f) Network plot showing relationships (proximity of nodes) and strengths of re-

lationships (line width between nodes).

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

Table 7

Node (item) parameters resulted from the network analysis.

Node Betweenness Closeness Strength Expected Influence

Item 2 0.44 0.45 −0.80 −1.01

Item 3 2.81 1.78 1.48 1.47

Item 4 0.89 1.40 1.59 1.59

Item 6 −0.47 0.69 0.21 0.23

Item 8 −0.58 −0.20 −0.51 −0.48

Item 9 −0.47 0.01 −0.27 −0.24

Item 12 1.34 −0.65 −0.31 −0.28

Item 16 −0.13 1.12 0.68 0.69

Item 18 0.44 1.24 2.32 2.30

Item 19 −1.15 −2.14 −1.66 −1.61

Item 20 0.10 0.74 −0.48 −0.44

Item 23 −0.36 −0.14 0.50 0.52

Item 24 −1.15 −0.61 −0.01 0.02

Item 27 −0.36 0.26 −0.44 −0.65

Item 30 −1.15 −1.45 −0.64 −0.61

Item 34 0.89 0.22 0.72 0.73

Item 35 0.89 −1.07 0.18 0.20

Item 36 −0.58 −0.51 −0.27 −0.24

Item 37 −0.36 −0.35 −0.79 −0.75

Item 38 −1.04 −0.80 −1.49 −1.44

11. General discussion

The purpose of the current paper was to develop and offer a prelimi-

nary validation for a measure focusing on climbing anxiety symptoms.

Our investigation contributes to the general literature on anxiety in

sports in several ways. First, we identified that anxiety in rock-

climbing, a high-risk sport unfolded into the same cognitive and affec-

tive facets as other existing measurement frameworks for anxiety in

sports, suggesting that the cognitive-affective model of anxiety general-

izes to the domain of high-risk sports (e.g., Martens et al., 1990). Sec-

ond, our findings offer preliminary evidence pertaining to the mea-

sure’s convergent and discriminant validity Third, in respect to crite-

rion-related validity, the newly developed measure accounted for vari-

ous performance-relevant outcomes in rock-climbing over generic mea-

sures of sport anxiety. Fourth, drawing from Item Response Theory, we

identified that the CAS-20 optimally differentiates between average,

above average and high levels of anxiety.

Taken together, our findings support the development of multi-

faceted, sport-specific anxiety measures, suggesting that employing

sport-specific measures might lead to an increased criterion-related va-

lidity. In respect to factorial validity, our investigation suggests that

climbing anxiety might be regarded as a three faceted construct, com-

prising elements of cognitive anxiety, somatic anxiety and self-

confidence. Theoretically, the multidimensionality of the climbing anx-

iety scale allows for a greater differentiation in understanding how each

dimension impacts performance or other relevant criteria. Taken to-

gether, our results suggest that anxiety in sport climbing might be suffi-

ciently different from existing sport-anxiety conceptualizations, war-

ranting a dedicated, contextualized measure. In respect to criterion va-

lidity, our measure accounted for performance-relevant outcomes over

existing measures of anxiety in sports and of climbing self-efficacy, sug-

gesting that the use of sport-specific item contents might capture con-

struct-relevant variance that further contributes to developing a better

understanding of anxiety’s role in explaining various sport and athletic-

relevant outcomes.

The IRT analysis shows good measurement characteristics for the

whole scale, with consistent information delivered by the total score,

especially between θ= 0 and θ= 3, with a peak of the information

function around z= 1.5 and the lowest standard error of measurement

between θ= 0 and θ= 2. This translates into the fact that the total

score is especially valid for anxiety levels that are average and above

average, up to about percentile 98. The items cover most of the range of

the construct well, albeit some items are not very strong in discrimina-

tive power; this translates into the fact that, while they contribute to a

meaningful total unidimensional model, some items could easily be ex-

cluded for a possible short form of the test. The network analysis identi-

fies the core indicators of the measured construct quite well (i18: ‘I

struggle to keep calm’) and also identifies three separate clusters of in-

dicators, corresponding to three different subscales that could give ex-

tra flavour in the interpretation of test-taker profiles.

From a theoretical standpoint the implications of our findings are in

line with previous research suggesting that employing sport-specific,

contextualized items to measure different facets of anxiety leads to a

higher criterion-related validity compared to measures that rely on gen-

eral, non-contextualized items. For example, in personality research,

extant research highlights the importance of using contextualized per-

sonality items to attain superior criterion-related validity in predicting

academic performance (Schmit et al., 1995) or various facets of job per-

formance (e.g., Bing et al., 2004;Holtz et al., 2005;Lievens et al.,

2008). Conceptually, this phenomenon was explained by the mecha-

nisms included in the Cognitive-Affective Personality System theory

(CAPS; Mischel & Shoda, 1995). CAPS posits that individual behaviours

depend on various situational specific cues and demands; ergo behav-

ioural consistencies are higher for situations having similar cues or con-

straints as opposed to situations imposing different demands on the in-

dividual. One of the unfolding measurement implications, generally re-

ferred to as frame of reference, is that the accuracy with which some

outcomes can be predicted increases when individuals respond to items

that provide them a specific contextual details or situational cues. The

lift in criterion-related validity when using contextualized items was ex-

plained via two different mechanisms. First, imposing a frame-of-

reference reduces the between-person variability in the way in which

individuals respond to various items included in the inventory. Second,

the use of contextualized items also reduces the within-person inconsis-

tencies, potentially impacting a measures’reliability and validity (e.g.,

Lievens et al., 2008). Considering that there are little differences in ob-

served reliabilities among the various anxiety measures employed in

our research, the most likely ‘culprit’for the lift in criterion-validity is

the reduced between-person inconsistencies as a consequence of con-

textualized sport climbing anxiety items. Additional research is war-

ranted to further explore this hypothesized explanation.

A potential explanation for our measure’s criterion-related validity,

the contextualized sport-specific anxiety measure accounted for vari-

ance in rock-climbing performance above and beyond the anxiety mea-

sure devised generically for sports (SAS-2) resides in overlap between

the frame of reference and the criterion domain. More specifically,

when responding to the climbing anxiety measure, the participants

adopt a frame of reference that overlaps with the criterion domain. In

line with the assumption that specific attitudes will have stronger rela-

tionships with specific behaviours, while general ones will contribute to

explaining broader outcomes (Fishbein & Ajzen, 1974), our findings

suggest that matching the level of specificity for anxiety measures to

the sport-specific performance criteria potentially leads to higher crite-

rion-related validity compared to employing measures that use generic,

non-sport specific items.

In respect to the practical implications, our findings indicate that

climbing anxiety affects various facets of sport climbing performance,

employing measured based on sport-specific frames of reference might

lead to more the Climbing Anxiety Scale, demonstrates its relevance in

several ways: (1) it predicts performance outcomes, over general and

even anxiety measures devised from sport contexts; (2) its validity in

accounting for performance outcomes remains intact even when con-

trolling for other variables such as self-efficacy; (3) it might have a

greater appeal for practitioners specializing in this athletic domain, for

two reasons - it includes constructs considered by experts to be relevant

and uses specific contents allowing a better reflection of the expression

of anxiety symptoms within this discipline and, implicitly enabling

practitioner to develop behaviourally targeted interventions; (4) Ar-

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

guably, especially for sports where either due to exposure or due to self-

selection (practitioners experiencing anxiety symptoms are unlikely to

even consider practicing the respective disciplines) contextualized mea-

sures are needed to accurately reflect the relationship between anxiety

and sport performance.

Limitations. Our investigation has several limitations. First, the use

of self-reported questionnaires collecting data about physiological man-

ifestations was criticized due to concerns about individuals' accuracy in

interpreting their symptoms (Woodman & Hardy, 2001). Second, crite-

rion-related validity should be established via longitudinal designs and

should draw on a sample of practitioners from several climbing disci-

plines (e.g., aid climbing, bouldering, deep water soloing, multi-pitch,

trad climbing, etc.). Third, to assess content validity, future studies

could employ an item-stems based approach, where series of items

could belong to specific item stems (e.g., ‘Usually, during a climb … ’).

Fourth, our sample was drawn from 18 different countries, and it was

not large enough to enable us to establish different levels of measure-

ment invariance for all the measures that we employed. Consequently,

potential biases related to cultural differences between the participants

could affect the current findings. Future studies should employ larger

samples of rock-climbers and thus, unlock the possibility to conduct in-

variance analyses as well. Fifth, unlike other sports, rock-climbing is a

less institutionalized discipline and participants are harder to identify

and to enrol. Consequently, it was not feasible at this initial develop-

ment stage to employ a different confirmation sample on which to ro-

bustly test the three-factor solution that we identified via the ex-

ploratory factorial analysis. Our proposed three-factor model should be

confirmed in a future replication sample. Sixth, another limitation is re-

lated to the potential impact of multicollinearity on accuracy and sta-

bility of the estimated parameters of the three factor CAS model. Multi-

collinearity could have resulted in artificially inflated standard errors,

imprecise parameter estimates and potentially undermining the true re-

lationships estimations amongst the observed variables. Consequently,

caution is advised when interpreting the factor structure, factor load-

ings and factor intercorrelations. Moreover, future research should

strive to establish a clear theory to explain necessity of using measures

that include sport-specific items, but otherwise tap into generalizable

constructs, such as anxiety. Future research endeavours should estab-

lish the utility of the CAS-20 in interventions and in working with elite/

professional climbers to gauge its applicability in this context. Future

research should also explore whether reliance on sport-specific items

leads to a more precise and valid measurement, increasing the validity

and utility of psychological measures in sports.

In conclusion, we contribute to the general domain of research in ex-

treme sports by developing a sport-specific anxiety measure and identi-

fying whether and how anxiety symptoms come into play in rock-

climbing performance. From a practical standpoint, the newly develop-

ment measure, CAS-20 can be used to assess anxiety in rock-climbers,

its validity in respect to climbing performance being higher than the

one observed for other sport anxiety or general anxiety measures. Fur-

thermore, practitioners and researchers alike can utilize the CAS-20 to

gain more in-depth understanding on how anxiety impacts various as-

pects of climbing performance. This understanding can inform the de-

velopment or enhancement of interventions. Additionally, it opens up

new avenues for understanding the intricate impact of anxiety in high-

risk sports.

CRediT authorship contribution statement

Maria Stefania Ionel: Writing –review & editing, Writing –origi-

nal draft, Visualization, Validation, Project administration, Methodol-

ogy, Investigation, Formal analysis, Data curation, Conceptualization.

Andrei Ion: Writing –review & editing, Writing –original draft, Visu-

alization, Validation, Supervision, Resources, Methodology, Formal

analysis, Data curation, Conceptualization. Dragos Iliescu: Writing –

review & editing, Writing –original draft, Visualization, Validation,

Supervision, Methodology, Formal analysis, Data curation. Laura

Visu-Petra: Writing –review & editing, Validation, Supervision, Re-

sources, Methodology, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal rela-

tionships which may be considered as potential competing interests:

Maria Stefania Ionel is collaborating with Climbing Psychology devel-

oping sport psychology services for climbers, and the International As-

sociation of Psychologists in Climbing (IAPSYC) and the International

Federation of Sport Climbing (IFSC) developing psychoeducation for

climbers.

Data availability

www.osf.io/wgv6e

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://

doi.org/10.1016/j.psychsport.2024.102635.

References

Aras, D., & Akalan, C. (2014). The effect of anxiety about falling on selected physiological

parameters with different rope protocols in sport rock climbing.The Journal of Sports

Medicine and Physical Fitness,54(1), 1–8. https://europepmc.org/article/med/

24445539.

Bing, M.N., Whanger, J.C., Davison, H.K., & VanHook, J.B. (2004). Incremental validity

of the frame-of-reference effect in personality scale scores: A replication and extension.

Journal of Applied Psychology,89(1), 150. https://doi.org/10.1037/0021-

9010.89.1.150.

Brown, T.A. (2015). Confirmatory factor analysis for applied research. Guilford

publications.

Browne, M.W., & Cudeck, R. (1992). Alternative ways of assessing model fit.Sociological

Methods & Research,21(2), 230–258. https://doi.org/10.1177/

0049124192021002005.

Brymer, E., Feletti, F., Monasterio, E., & Schweitzer, R. (2020). Understanding extreme

sports: A psychological perspective.Frontiers in Psychology,10, 3029. https://doi.org/

10.3389/fpsyg.2019.03029.

Child, D. (2006). The essentials of factor analysis. A&C Black.

Clough, P., Houge Mackenzie, S., Mallabon, L., & Brymer, E. (2016). Adventurous

physical activity environments: A mainstream intervention for mental health.Sports

Medicine,46, 963–968. https://doi.org/10.1007/s40279-016-0503-3.

Cooke, A., Kavussanu, M., McIntyre, D., & Ring, C. (2011). Effects of competition on

endurance performance and the underlying psychological and physiological mechanisms.

Biological Psychology,86(3), 370–378. https://doi.org/10.1016/

j.biopsycho.2011.01.009.

Cox, R.H., Martens, M.P., & Russell, W.D. (2003). Measuring anxiety in athletics: The

revised competitive state anxiety inventory–2.Journal of Sport & Exercise Psychology,25

(4), 519–533. https://doi.org/10.1123/jsep.25.4.519.

DeWolfe, C.E., Galbraith, M.K., Smith, M.M., Watt, M.C., Olthuis, J.V., Sherry, S.B., &

Stewart, S.H. (2023). Anxiety sensitivity and physical activity are inversely related: A

meta-analytic review.Mental Health and Physical Activity,25, 100548. https://doi.org/

10.1016/j.mhpa.2023.100548.

Draper, N., Dickson, T., Fryer, S., Blackwell, G., Winter, D., Scarrott, C., & Ellis, G.

(2012). Plasma cortisol concentrations and perceived anxiety in response to on-sight rock

climbing.International Journal of Sports Medicine,33(1), 13–17. https://doi.org/

10.1055/s-0031-1284348.

Draper, N., Giles, D., Schöffl, V., Konstantin Fuss, F., Watts, P., Wolf, P., …Abreu, E.

(2015). Comparative grading scales, statistical analyses, climber descriptors and ability

grouping: International Rock Climbing Research Association position statement.Sports

Technology,8(3–4), 88–94. https://doi.org/10.1080/19346182.2015.1107081.

Fishbein, M., & Ajzen, I. (1974). Attitudes towards objects as predictors of single and multiple

behavioral criteria.Psychological Review,81(1), 59. https://doi.org/10.1037/

h0035872.

Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable

variables and measurement error.Journal of Marketing Research,18(1), 39–50. https://

doi.org/10.2307/3151312.

Garrido-Palomino, I., & España-Romero, V. (2023). Fear of falling in women: A

psychological training intervention improves climbing performance.Journal of Sports

Sciences, 1–12. https://doi.org/10.1080/02640414.2023.2281157.

Gouttebarge, V., Castaldelli-Maia, J.M., Gorczynski, P., Hainline, B., Hitchcock, M.E.,

Kerkhoffs, G.M., …Reardon, C.L. (2019). Occurrence of mental health symptoms and

disorders in current and former elite athletes: A systematic review and meta-analysis.

British Journal of Sports Medicine,53(11), 700–706. https://doi.org/10.1136/bjsports-

CORRECTED PROOF

M.S. Ionel et al. Psychology of Sport & Exercise xxx (xxxx) 102635

2019-100671.

Hodgson, C.I., Draper, N., McMorris, T., Jones, G., Fryer, S., & Coleman, I. (2009).

Perceived anxiety and plasma cortisol concentrations following rock climbing with differing

safety rope protocols.British Journal of Sports Medicine,43(7), 531–535. https://

doi.org/10.1136/bjsm.2007.046011.

Holtz, B.C., Ployhart, R.E., & Dominguez, A. (2005). Testing the rules of justice: The effects

of frame-of-reference and pre-test validity information on personality test responses and test

perceptions.International Journal of Selection and Assessment,13(1), 75–86. https://

doi.org/10.1111/j.0965-075X.2005.00301.x.

Horn, J.L. (1965). A rationale and test for the number of factors in factor analysis.

Psychometrika,30, 179–185. https://www.j-humansciences.com/ojs/index.php/

IJHS/article/view/1674.

Hu, L.T., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure

analysis: Conventional criteria versus new alternatives.Structural Equation Modeling: A

Multidisciplinary Journal,6(1), 1–55. https://doi.org/10.1080/10705519909540118.

Hunsley, J., & Meyer, G.J. (2003). The incremental validity of psychological testing and

assessment: Conceptual, methodological, and statistical issues.Psychological Assessment,

15(4), 446. https://doi.org/10.1037/1040-3590.15.4.446.

Hunt, J.C. (1995). Divers’accounts of normal risk.Symbolic Interaction,18(4), 439–462.

https://doi.org/10.1525/si.1995.18.4.439.

IBM Corp. (2019). IBM SPSS statistics for windows, version 26.0. Armonk, NY: IBM Corp.

Janelle, C.M., Fawver, B.J., & Beatty, G.F. (2020). Emotion and sport performance. In

G., Tenenbaum, R.C., Eklund, & N., Boiangin (Eds.), Handbook of sport psychology:

Social perspectives, cognition, and applications (4th ed., pp. 254–298). John Wiley &

Sons, Inc. https://doi.org/10.1002/9781119568124.ch13.

Jobst, L., Bader, M., & Moshagen, M. (2023). A tutorial on assessing statistical power and

determining sample size for structural equation models.Psychological Methods,28,

207–221. https://doi.org/10.1037/met0000423.

Johnston, R., Jones, K., & Manley, D. (2018). Confounding and collinearity in regression

analysis: A cautionary tale and an alternative procedure, illustrated by studies of British

voting behaviour.Quality and Quantity,52, 1957–1976. https://doi.org/10.1007/

s11135-017-0584-6.

Jones, E.S., Mullen, R., & Hardy, L. (2019). Measurement and validation of a three factor

hierarchical model of competitive anxiety.Psychology of Sport and Exercise,43, 34–44.

https://doi.org/10.1016/j.psychsport.2018.12.011.

Liebert, R.M., & Morris, L.W. (1967). Cognitive and emotional components of test anxiety: A

distinction and some initial data.Psychological Reports,20(3), 975–978. https://doi.org/

10.2466/pr0.1967.20.3.975.

Lievens, F., De Corte, W., & Schollaert, E. (2008). A closer look at the frame-of-reference

effect in personality scale scores and validity.Journal of Applied Psychology,93(2), 268.

https://doi.org/10.1037/0021-9010.93.2.268.

Llewellyn, D.J., Sanchez, X., Asghar, A., & Jones, G. (2008). Self-efficacy, risk taking and

performance in rock climbing.Personality and Individual Differences,45(1), 75–81.

https://doi.org/10.1016/j.paid.2008.03.001.

Lovibond, P.F., & Lovibond, S.H. (1995). The structure of negative emotional states:

Comparison of the depression anxiety stress scales (DASS) with the beck depression and

anxiety inventories.Behaviour Research and Therapy,33(3), 335–343. https://doi.org/

10.1016/0005-7967(94)00075-U.

Martens, R. (1977). Sport competition anxiety test. Champaign, IL: Human Kinetics.

Martens, R., Vealey, R.S., & Burton, D. (1990). Competitive anxiety in sport. Champaign,

IL: Human Kinetics.

Masked, 2022.

McDonald, R.P. (1999). Test theory: A unified treatment. Lawrence Erlbaum Associates

Publishers.

Mischel, W., & Shoda, Y. (1995). A cognitive-affective system theory of personality:

Reconceptualizing situations, dispositions, dynamics, and invariance in personality

structure.Psychological Review,102(2), 246. https://doi.org/10.1037/0033-

295X.102.2.246.

Mullen, R., Hardy, L., & Tattersall, A. (2005). The effects of anxiety on motor performance:

A test of the conscious processing hypothesis.Journal of Sport & Exercise Psychology,27

(2), 212–225. https://doi.org/10.1123/jsep.27.2.212.

Muraki, E. (1992). A generalized partial credit model: Application of an EM algorithm.Applied

Psychological Measurement,16(2), 159–176. https://doi.org/10.1002/j.2333-

8504.1992.tb01436.x.

Pijpers, J.R., Oudejans, R.R., Bakker, F.C., & Beek, P.J. (2006). The role of anxiety in

perceiving and realizing affordances.Ecological Psychology,18(3), 131–161. https://

doi.org/10.1207/s15326969eco1803_1.

Pijpers, J.R., Oudejans, R.R., Holsheimer, F., & Bakker, F.C. (2003). Anxiety–performance

relationships in climbing: A process-oriented approach.Psychology of Sport and Exercise,4

(3), 283–304. https://doi.org/10.1016/S1469-0292(02)00010-9.

R Core Team. (2021). R: A language and environment for statistical computing. Vienna,

Austria: R Foundation for Statistical Computing. http://www.R-project.org.

Reardon, C.L., Hainline, B., Aron, C.M., Baron, D., Baum, A.L., Bindra, A., …

Engebretsen, L. (2019). Mental health in elite athletes: International Olympic Committee

consensus statement (2019).British Journal of Sports Medicine,53(11), 667–699.

https://doi.org/10.1136/bjsports-2019-100715.

Robinson, D.W. (1985). Stress seeking: Selected behavioral characteristics of elite rock

climbers.Journal of Sport & Exercise Psychology,7(4), 400–404. https://doi.org/

10.1123/jsp.7.4.400.

Schmit, M.J., Ryan, A.M., Stierwalt, S.L., & Powell, A.B. (1995). Frame-of-reference

effects on personality scale scores and criterion-related validity.Journal of Applied

Psychology,80(5), 607. https://doi.org/10.1037/0021-9010.80.5.607.

Smith, R.E., Smoll, F.L., Cumming, S.P., & Grossbard, J.R. (2006). Measurement of

multidimensional sport performance anxiety in children and adults: The Sport Anxiety

Scale-2.Journal of Sport & Exercise Psychology,28(4), 479–501. https://doi.org/

10.1123/jsep.28.4.479.

Smith, R.E., Smoll, F.L., & Schutz, R.W. (1990). Measurement and correlates of sport-

specific cognitive and somatic trait anxiety: The Sport Anxiety Scale.Anxiety Research,2

(4), 263–280. https://doi.org/10.1080/08917779008248733.

Tabachnick, B.G., & Fidell, S.L. (2007). Using multivariate statistics (5th ed.). Allyn &

Bacon/Pearson Education. https://psycnet.apa.org/record/2006-03883-000.

Teixeira, D.S., Rodrigues, F., Monteiro, D., & Cid, L. (2022). The Behavioral Regulation in

Exercise Questionnaire (BREQ-4): Psychometric evidence of introjected approach

regulation in Portuguese health club exercisers.Psychology of Sport and Exercise,63,

102286. https://doi.org/10.1016/j.psychsport.2022.102286.

Waltz, C.F., & Bausell, R.B. (1983). Nursing research: Design, statistics and computer

analysis (2nd ed.). Philadelphia: FA Davis Company.

Woodman, T., & Hardy, L. (2001). Stress and anxiety. In R.N., Singer, H.A., Hausenblas,

& C.M., Janelle (Eds.), Handbook of sport psychology (2nd ed., pp. 290–318). New

York: Wiley.

ResearchGate has not been able to resolve any citations for this publication.

Fear of falling in women: A psychological training intervention improves climbing performance View supplementary material Fear of falling in women: A psychological training intervention improves climbing performance

Article

Full-text available

Nov 2023
J SPORT SCI

The main purpose of this study was to determine the effect of a psychological training intervention based on emotional regulation on anxiety and climbing ability in women climbers with fear of falling. A secondary aim was to compare the outcomes of climbing ability, anxiety, self-confidence and inter-oceptive awareness (IA) between the psychological group (PG), a training (TG) and a control group. Self-reported climbing ability, anxiety and IA were assessed using Competitive State Anxiety Inventory-2 and the Multidimensional Assessment of Interoceptive Awareness questionnaire. Results indicated a significant improvement in climbing ability for both PG and TG. The PG showed a significant reduction in cognitive and somatic anxiety, while the TG only exhibited a reduction in cognitive anxiety. Moreover, the PG demonstrated a greater significant increase in self-confidence compared to the TG. IA improved in five subscales for the PG, whereas the TG changed in only two subscales. In conclusion, a psychological training intervention focusing on emotional regulation might contribute to improvements in IA, reduced anxiety levels, and enhancements in climbing ability and self-confidence among women climbers facing fear of falling. ARTICLE HISTORY

Personality, grit, and performance in rock-climbing: down to the nitty-gritty

Article

Full-text available

Mar 2023

Over the past decades, the study of personality gained momentum in the field of sport psychology, a prime example being the increased number of publications relating personality to performance in extreme sports. The aim of the present investigation was twofold: to relate Five-Factor Model (FFM) personality traits with various aspects of climbing performance and to estimate whether grit, as a distinct personality trait, predicts climbing performance over and beyond FFM dimensions. Our sample included adult bouldering and sport climbing practitioners (N = 272 sport climbers, 155 boulderers) with ages between 16 and 69 (M = 32.1, SD = 10.0). We measured personality (Big Five Inventory FFM–2 Short Form, and a 12-item Grit Scale) and multiple indicators of outdoor sport climbing and bouldering performance. Results indicated that climbing performance was predicted by openness and agreeableness. Grit also significantly predicted climbing performance over the FFM traits. Contrary to the commonly held view depicting grit as indistinct from conscientiousness, our findings suggest that this trait has a unique contribution to explaining performance in a relatively novel, high-risk sport.

A Tutorial on Assessing Statistical Power and Determining Sample Size for Structural Equation Models

Article

Full-text available

Oct 2021

Structural equation modeling (SEM) is a widespread approach to test substantive hypotheses in psychology and other social sciences. However, most studies involving structural equation models neither report statistical power analysis as a criterion for sample size planning nor evaluate the achieved power of the performed tests. In this tutorial, we provide a step-by-step illustration of how a priori, post hoc, and compromise power analyses can be conducted for a range of different SEM applications. Using illustrative examples and the R package semPower, we demonstrate power analyses for hypotheses regarding overall model fit, global model comparisons, particular individual model parameters, and differences in multi-group contexts (such as in tests of measurement invariance). We encourage researchers to yield reliable – and thus more replicable – results based on thoughtful sample size planning, especially if small or medium-sized effects are expected.

Emotion and Sport Performance

Article

Full-text available

Jun 2020

This chapter provides working definitions of emotion and related concepts utilized in the study of emotions and sport performance. It overviews multiple important roles that emotions serve in the resultant behaviors that ultimately influence sport performance. The chapter reviews influential conceptual frameworks that have advanced understanding of emotions as related to sport performance. It highlights empirical work that has contributed deductively to theory testing and inductively to the advancement of knowledge. The integrative model of stress, attention, and human performance incorporates, among others, hypotheses from the attention control theory. The chapter also highlights how emotions alter the fundamental mechanisms underpinning behavioral changes that impact sport performance. It offers pragmatic consideration of how sport psychology research can continue to inform the scientific practice of sport psychology, particularly as related to the development of innovative approaches for emotion regulation.

Editorial: Understanding Extreme Sports: A Psychological Perspective

Article

Full-text available

Jan 2020

Mental health in elite athletes: International Olympic Committee consensus statement (2019)

Article

Full-text available

May 2019
BRIT J SPORT MED

Mental health symptoms and disorders are common among elite athletes, may have sport related manifestations within this population and impair performance. Mental health cannot be separated from physical health, as evidenced by mental health symptoms and disorders increasing the risk of physical injury and delaying subsequent recovery. There are no evidence or consensus based guidelines for diagnosis and management of mental health symptoms and disorders in elite athletes. Diagnosis must differentiate character traits particular to elite athletes from psychosocial maladaptations. Management strategies should address all contributors to mental health symptoms and consider biopsychosocial factors relevant to athletes to maximise benefit and minimise harm. Management must involve both treatment of affected individual athletes and optimising environments in which all elite athletes train and compete. To advance a more standardised, evidence based approach to mental health symptoms and disorders in elite athletes, an International Olympic Committee Consensus Work Group critically evaluated the current state of science and provided recommendations.

Occurrence of mental health symptoms and disorders in current and former elite athletes: A systematic review and meta-analysis

Article

Full-text available

May 2019
BRIT J SPORT MED

Objectives To present an overview of the existing epidemiological evidence regarding the occurrence of mental health symptoms and disorders among current and former elite athletes. Design Systematic review and meta-analysis. Data sources Five electronic databases were searched from inception to November 2018: PubMed (MEDLINE), SportDiscus via EBSCO, PSycINFO via ProQuest, Scopus and Cochrane. Eligibility criteria for selecting studies We included original quantitative studies that were written in English, were conducted exclusively among current or former elite athletes, and presented incidence or prevalence rates of symptoms of mental disorders. Results Twenty-two relevant original studies about mental health symptoms and disorders among current elite athletes were included: they presented data especially on symptoms of distress, sleep disturbance, anxiety/depression and alcohol misuse. Meta-analyses comprising 2895 to 5555 current elite athletes showed that the prevalence of mental health symptoms and disorders ranged from 19% for alcohol misuse to 34% for anxiety/depression. Fifteen relevant original studies about mental health symptoms and disorders among former elite athletes were included: they similarly presented data especially about symptoms of distress, sleep disturbance, anxiety/depression and alcohol misuse. Meta-analyses comprising 1579 to 1686 former elite athletes showed that the prevalence of mental health symptoms and disorders ranged from 16% for distress to 26% for anxiety/depression. Conclusions Our meta-analyses showed that the prevalence of mental health symptoms and disorders ranged from 19% for alcohol misuse to 34% for anxiety/depression for current elite athletes, and from 16% for distress to 26% for anxiety/depression for former elite athletes.

Anxiety sensitivity and physical activity are inversely related: A meta-analytic review

Article

Aug 2023

The Behavioral Regulation in Exercise Questionnaire (BREQ-4): Psychometric Evidence of Introjected Approach Regulation in Portuguese Health Club Exercisers

Article

Sep 2022
PSYCHOL SPORT EXERC

The assessment of motivation has been a key aspect to the understanding of exercise participation, and research grounded in self-determination theory has presented valid and reliable instruments for that purpose. Given the need to continually refine this latent construct, the present study aimed to translate, adapt, and psychometrically validate a subscale targeting the approach facet of introjection, and to test the pattern of associations between motives for practice, basic psychological needs satisfaction/frustration, and behavioral regulations encompassing the validated introjection subscale, in a sample of health club exercisers. For that purpose, two studies were developed with a total of 1216 health club exercisers. In Study I (n = 806), Confirmatory Factor Analysis and Exploratory Structural Equation Modeling analysis to test the motivational continuum encompassing the introjected approach subscale were performed. In Study II (n = 410) associations and structural models between intrinsic and extrinsic goal contents, basic psychological needs satisfaction and frustration, and behavioral regulations with the new subscale were tested. The correlated seven-factor model with 21 items in Study I displayed good psychometric properties (CFA: χ² = 481.977 (168), p <.001, CFI = .936, TLI = .915, SRMR = .037, RMSEA = .048; ESEM: χ² = 178.672 (84), p <.001, CFI = .980, TLI = .949, SRMR = .014, RMSEA = .037). The introjected approach regulation added to the preexisting factorial structure did not affect the validity and reliability of the instrument. The results from Study II supported a theoretically expected pattern of associations, in which the introjected regulation of approach is positioned between introjected avoidance and identified regulation along the motivational continuum. Additionally, path estimates depicted criterion validity for the new subscale. All in all, this work presents preliminary evidence for an introjected approach regulation subscale that can be used in health club practices for a better understanding of the motivational quality of exercise practice.

Test Theory: A Unified Treatment

Book

Jun 2013

Roderick P. McDonald

Climbing anxiety scale (CAS-20): Preliminary development and validation

Abstract

Recommended publications

Personality, grit, and performance in rock-climbing: down to the nitty-gritty

Personality and rock-climbing performance progression: a 12-months follow-up

Role of emotional intelligence on rock climbing performance. [El rol de la inteligencia emocional en...

Personality, Grit, and Performance in Rock-climbing: Down to the Nitty-Gritty