Cette thèse propose un test empirique du modèle d’attribution du leadership basé sur l’observation de dyades leader-subordonné au Sénégal et en France. Ce modèle se concentre sur des construits essentiels comme la complexité attributionnelle et l’exactitude des attributions des leaders, les comportements qui en résultent, les variables médiatrices comme la satisfaction des subordonnés, les stratégies correctives de la performance et les variables de résultat comme les perceptions du leadership. Le modèle est testé à l’aide de régressions hiérarchiques des effets directs et des effets médiateurs des attributions biaisées. Notre intention est de tester les relations à partir d’un échantillon d’auditeurs en formation continue de l’enseignement supérieur (au Sénégal et en France) et de déterminer selon notre modèle et nos hypothèses de recherche, quels construits contribuent le plus à expliquer la complexité attributionnelle des leaders. Cequi nous amène à étudier l’influence directe de l’exactitude des attributions, l’effet des stratégies correctives et de la satisfaction des subordonnés sur la perception du leadership. Nous testerons aussi le rôle des attributions biaisées sur les deux variables médiatrices. Les hypothèses sont fortement corroborées. La complexité attributionnelle des leaders est liée à l’exactitude de leurs attributions, telle qu’elle est perçue par leurs subordonnés. Les stratégies de correction de la performance élaborées par les leaders se sont révélées liées à des attributions exactes, puis à des variables clés de résultats. Nous analysons les résultats ainsi que les limites et les orientations futures de la recherche. Les contributionsmanagériales sont doubles : d'une part, de prendre en compte dans le processus de recrutement des leaders leur aptitude à adopter une approche attributionnelle complexe et, d'autre part, les former afin de mieux accompagner leurs subordonnés. / This thesis proposes an empirical test of the leadership attribution model based on the observation of leader-subordinate dyads in Senegal and France. This model focuses on key constructs such as attributional complexity and leader attribution accuracy, resulting behaviors, mediating variables such as subordinate satisfaction, performance corrective strategies, and outcome variables such as leadership perceptions. The model is tested using hierarchical regressions of direct effects and mediating effects of biased allocations. Our intention is to test the relationships from a sample of auditors in continuingeducation of higher education (in Senegal and France) and to determine according to our model and our research hypotheses, which constructs contribute the most to explain the attributional complexity of the leaders. This leads us to study the direct influence of the accuracy of the attributions, the effect of the corrective strategies and the satisfaction of the subordinates on the perception of the leadership. We will also test the role of biased assignments on the two mediating variables. Hypotheses are strongly corroborated. The attributional complexity of leaders is linked to the accuracy of their attributions, as perceived by their subordinates. The performance correction strategies developed by leaders have beenlinked to accurate attribution and then to key outcome variables. We analyze the results as well as the limits and future directions of the research. The managerial contributions are twofold: on the one hand, to take into account in the process of recruiting leaders their ability to adopt a complex attributional approach and, on the other hand, to train them to better support their subordinates.

Summaries in English, Tsonga and Venda / Dyondzo yi lavisisa vutlhari lebyi nga kona eka marungula ya tiprosi ta Xitsonga byo kambisisa swikili swa vulawuri na switirateji swa mbulavurisano leswi tirhisiwaka hi varhangeri va sweswi eka tisosayiti. Mahungu hinkwawo ya mbulavurisano wa swiharhi na vanhu ya nyikiwile lama kongomisaka eka tinxaka ta vurhangeri na swikili. Marungula ya tiprosi ta Xitsonga lama hlawuriweke na ku dokhumentiwa na marungula hi varhangeri va sweswi ya hlengeletiwile naswona ya xopaxopiwile. Dyondzo yi leteriwile hi Nxopanxopo wa Mbulavulo wa Nkoka, Mbulavurisano wa Vanhu, na maendlelo ya vutitivisi bya Vanhu. Datara yi lulamisiwile hi mitlawa ku ya hi mikhetekanyo leyi hlawuriweke ya switirateji swa mbulavurisano wa nomu na wo ka wu nga ri wa nomu. Mavulavulelo ya vanhu lava nga eka swiyimo swa vurhangeri eka marungula ya tiprosi na ya vanhu va sweswi ya hlamuseriwile na ku va ya xopaxopiwile hi ku ya hi maendlelo lama hlawuriweke. Leswi nga kumiwa swi humeserile erivaleni leswaku hakunene, varhangeri va sweswi va kayivela swikili swa vulawuri na switirateji swa mbulavurisano. Milawu ya vurhangeri leyi kambisisiweke leyi languteriweke ku suka eka varhangeri votala va vanhu yi kumiwile yi nga ri kona. Swiyimo swa malawulelo na mafumelo yo biha hi ndlela ya ntirhiso wo ka wu nga ri kahle kumbe nkucetelo swi kombisiwile ku suka eka ku tirhisa switirateji swa mbulavurisano wa nomu na wo ka wu nga ri wa nomu hi ndlela yo ka yi nga ri kahle. Makolo, vukari na ku tshembhela eka swa vukhongeri kumbe swikwembu swa le hansi a ku ri swona a swi vanga malawulelo na mafumelo yo biha. Vatirhivotala va xiyimo xa le hansi va hlanganile na ku xanisiwa emimoyeni, ku xanisiwa hi ku vulavula na ku xanisiwa emirini, ku xanisiwa hi swa masangu na ku susiwa eka swiyimo swa vona ehansi ka vulawuri bya varhangeri va vona eka sosayiti. Varhangeri va sweswi va nga dyondza swotala hi swikili swa vulawuri na switirateji swa mbulavurisano ku suka eka marungula ya tiprosi ta tindzimi ta Afrika, ngopfu-ngopfu marungula ya tiprosi ta Xitsonga. Dyondzo yi humesela erivaleni leswaku ntsheketo wu fuwile hi vutlhari bya swikwembu swa le hansi, lebyi nga pfunaka eka swiphiqo swa vurhangeri bya sweswi. Dyondzo yi nga va na nkoka eka varhangeri na vatirhikulorhi va vona va xiyimo xa le hansi. Swiringanyeto na swibumabumelo swi nyikiwile eka varhangeri va sweswi ku va pfuna ku teka xiboho xo hundzuluxa switirateji swa vona swa mbulavurisano na ku cinca mavulavulelo ya vona. Varhangeri va sweswi va fanele va kopela ku suka eka vutlhari lebyi nga kona eka marungula ya tiprosi ku kambisisa ku kayivela ka swikili swa vona swa vulawuri na switirateji swa mbulavurisano ku antswisa mbulavurisano exikarhi ka vona na vatirhikulorhi va vona lava nga eka xiyimo xa le hansi. / Ngudo i todisisa vhutali vhu re hone kha nganetshelo dza phurosa ya sialala ya Xitsonga malugana na khaedu ya zwikili zwa ndangulo na ndila dza vhudavhidzani zwi shumiswaho nga vharangaphanda vha musalauno zwitshavhani. Mafhungo a u rangela malugana na vhudavhidzani ha phukha na ha vhathu o netshedzwa ho sedzwa maitele na zwikili zwa vhurangaphanda. Nganetshelo dzo vhulungwaho na u nanguludzwa dza phurosa ya sialala ya Xitsonga na nganetshelo malugana na vharangaphanda vha musalauno dzo kuvhanganywa dza senguluswa. Ngudo iyi yo disendeka kha ndila dza Tsenguluso ya Vhudavhidzani ha Tsasaladzo, Ngudo ya luambo ho sedzwa zwa u Tshilisana kha Vhudavhidzani na Nzudzanyele ya Vhushaka ha vhathu. Data yo vhekanywa u ya nga zwigwada zwa ndila dza vhudavhidzani dza u amba na dzi si dza u amba. Maitele a u davhidzana a vhaanewa kha vhuimo ha vhurangaphanda kha nganetshelo dza phurosa dza sialala na kha zwitshavha zwa musalauno o talutshedzwa a senguluswa ho d isendekwa nga ndila dzo nangwaho. Mawanwa o dzumbulula uri nangoho, vharangaphanda vha musalauno vha shaya zwikili zwa ndangulo na ndila dza vhudavhidzani. Mirando ya vhurangaphanda yo tolwaho ine ya lavhelelwa kha vharangaphanda vhanzhi vha zwitshavha a vha nayo. Zwiwo zwa ndangulo na u laula zwa u shumisa maanda kha tshivhumbeo tsha u shumisa kana u tutuwedza hu sa vhuedzi zwo vhonala kha tshumiso i si yone ya ndila dza vhudavhidzani dza u amba na dzi si dza u amba. Tseda, tsinyuwo na lutendo lwa vhurereli vhu si hone kana vha kale zwo vha zwiko zwa ndangulommbi na u toda u laula. Vhunzhi ha vhalanda vho vhaisala muyani, kha u amba na u tambudzwa muvhilini, vho tambudzwa kha zwa vhudzekani na u sudzuluswa nga vharangaphanda vhavho tshitshavhani. Vharangaphanda vha musalauno vha nga guda zwinzhi nga ha zwikili zwa ndangulo na ndila dza vhudavhidzani kha nganetshelo dza phurosa ya sialala ya nyambo dza vharema, zwihulu kha nganetshelo dza phurosa ya sialala ya Xitsonga. Ngudo yo bvukulula uri zwa sialala zwo pfuma nga vhutali ha vha kale, vhune vhu nga thusa malugana na thaidzo dza vhurangaphanda ha musalauno. Ngudo iyi i nga vha ya ndeme kha vharangaphanda na vhalanda vhavho. Madzinginywa na themendelo zwo netshedzwa kha vharangaphanda vha musalauno u vha thusa uri vha kone u tshea malugana na u dzudzanya nga huswa ndila dzavho dza vhudavhidzani, nahone vha kone u shandukisa kuitele kwavho kwa u davhidzana. Vharangaphanda vha muisalauno vha fanela u guda kha vhutali vhu re kha nganetshelo dza phurosa ya sialala vha lulamise tshiimo tsha u shaya havho zwikili zwa ndangulo na nd ila dza vhudavhidzani malugana na u khwinisa vhudavhidzani vhukati havho na vhalanda vhavho. / The study investigates the wisdom embedded in Xitsonga oral prose narratives to address management skills and communication strategies used by current leaders in societies. Background information on animal and human communication is provided with a focus on leadership styles and skills. Selected documented Xitsonga oral prose narratives and narratives about current leaders were collected and analysed. The study is guided by the Critical Discourse Analysis, Interactional Sociolinguistics and Social Constructionist approaches. Data was classified per the selected categories of the verbal and nonverbal communication strategies. Communicative behaviours of characters in leadership positions in the oral prose narratives and those of the current societies were interpreted and analysed based on the selected approaches. Findings revealed that indeed, current leaders lack management skills and communication strategies. Examined leadership principles expected from many societal leaders were found to be non-existent. Cases of power mismanagement and dominance in the form of negative manipulation or influence were identified from their misuse of verbal and nonverbal communication strategies. Greed, anger and defective religious or ancestral beliefs were sources of such power mismanagement and dominance. Most subordinates suffered emotional, verbal and physical abuse, sexual harassment and displacement at the hands of their leaders in the society. Current leaders can learn much about management skills and communication strategies from the African languages oral prose narratives, particularly Xitsonga oral prose narratives. The study reveals that folklore is rich with ancestral wisdom, which can assist in current leadership problems. The study can be valuable to both leaders and their subordinates. Suggestions and recommendations are offered to the current leaders to help them decide on adjusting their communication strategies and ultimately changing their communicative behaviour. Current leaders should take a leaf from the wisdom embedded in oral prose narratives to address their lack of management skills and communication strategies to improve communication between them and their subordinates. / African Languages / D. Litt. et Phil. (African Languages)