Examining the importance of supporting youth’s basic needs in one youth leadership programme: a case study exploring programme quality

Abstract Researchers outline the importance of understanding features that enhance quality within youth programmes. One feature is youth’s basic psychological needs support. The purpose of this study was to examine if differences existed between two leaders in delivery of one youth leadership programme related to programme quality and needs support. A mixed-methods approach was employed. Programme quality was assessed quantitatively from researchers' observations. Youth self-reported on needs support and participated in semi-structured interviews. Results indicated significant differences on programme quality with Leader 1 rating higher on safe and supportive environment than Leader 2. Youth perceived Leader 1 as more effective than Leader 2 in providing a psychologically safe environment and fostering needs support. Implications and future directions are discussed.

exploring how one's role as a leader impacted their development. Findings indicated that the leaders believed their participation helped them develop leadership-related skills such as becoming more responsible for themselves, becoming more confident, learning how to communicate effectively to both youth and adults and learning how to problem solve and manage stressful situations. Although ample research has been conducted to investigate the outcomes of participation in leadership programmes, little research has examined the factors that foster or hinder such outcomes within youth programmes.
Programme quality has been identified as a critical factor that has been posited as the best predictor of positive developmental outcomes in youth (Durlak et al., 2010;Vandell, Pierce, & Dadisman, 2004). A number of common features have been outlined that can contribute to programme quality. For example, Petitpas, Cornelius, Van Raalte and Jones (2005) developed a framework that outlines four specific components critical to the fostering of youth development: (a) context, (b) external assets, (c) internal assets and (d) research and evaluation. First, the researchers assert that the context of a programme involves creating a psychologically safe environment where participants can develop a sense of initiative, can find a valued role within a group, and where involvement is voluntary. External assets involve being surrounded by responsible and caring adult mentors and a positive peer group. The programme must also involve intentional teaching of life skills (internal assets) youth need to develop to succeed. Lastly, Petitpas and colleagues (2005) outlined that a formal programme evaluation needs to be incorporated. Programme evaluations have the ability to provide evidence-based and strategic programme planning for the organization in the future (Danish, Forneris, Hodge, & Heke, 2004). Petitipas et al., (2005) purported that case study research may be a beneficial methodology so that programmes can evolve to best meet the needs of the participants. Similarly, Lerner (2004) proposed 'the Big Three' , which suggests three strategies to be incorporated into youth programmes: (a) positive and sustained adult-youth relations, (b) youth life-skill building activities and (c) youth participation and leadership in community activities. Finally, the National Research Council and Institute of Medicine (NRCIM) proposed eight strategies for enhancing programme quality that can lead to youth development: (1) physical and psychological safety; (2) appropriate structure; (3) supportive relationships; (4) opportunities to belong; (5) positive social norms; (6) support for efficacy and mattering; (7) opportunities for skill building; (8) integration of family, school and community efforts (Eccles & Gootman, 2002). For the purposes of this research, programme quality was measured based on Eccles and Gootman's work, as these programme quality setting features have been widely recognized and researched within the youth development literature.
Recent research has also examined the role of basic needs support in fostering youth development. Youth programmes are more effective at enhancing development when youth have the opportunity to build supportive relationships with adult leaders (Armour & Sandford, 2013;Eccles & Gootman, 2002;Roth & Brooks-Gunn, 2003). The Basic Needs Theory (BNT), a sub-theory within Self-Determination Theory (Deci & Ryan, 1985), proposes that humans function effectively as a result of the social environment and its potential for supporting three basic needs (Deci & Vansteenkiste, 2004;Ryan & Deci, 2000a, 2000b. Deci (2000b, 2002) argue that autonomy, competence and relatedness are three universal psychological needs that are essential for growth, integrity and wellness. Autonomy involves feeling internal assent regarding one's behaviour; feelings of choice, rather than feeling controlled or pressured; competence involves feeling efficient, effective and masterful in one's behaviour, rather than incompetent and ineffective; and relatedness involves feeling meaningfully connected to others, rather than feeling alienated or ostracized. Conditions that support an individual's experience of autonomy, competence and relatedness are argued to foster high quality forms of motivation and engagement for activities, leading to positive psychological development (Ryan & Deci, 2000b, 2002. Environments that do not support one or all of these needs have been postulated to undermine well-being. Therefore, social environments that support or thwart these basic needs can have a positive or detrimental influence on one's well-being in that setting. As mentioned above, a number of researchers have begun to study basic needs in youth contexts, specifically examining the role of autonomy (e.g. Adie, Duda, & Ntoumanis, 2012;Conroy & Douglas Coatsworth, 2007;Standage, Duda, & Pensgaard, 2005); however, it has been argued that support of all three needs is required for internalization (Deci & Ryan, 2011).
For example, Quested and Duda (2010) found that perceptions of autonomy support were positively related to autonomy and relatedness and that needs satisfaction positively predicted positive affect. Yohalem and Wilson-Ahlstrom (2010) highlighted that not only has there been increased emphasis placed on the importance of programme quality, there is also growing evidence that high quality programmes can affect a range of youth outcomes. Moreover, the support of basic needs are believed to lead to greater psychosocial outcomes (e.g. Adie et al., 2012;Hodge, Danish, & Martin, 2013). In contrast, when programme quality is lower, lower perceptions of needs support may occur. For example, a study conducted by Bean and Forneris (submitted) examined programme quality and basic needs support in two in-school mentoring programmes and found that higher scores of programme quality predicted higher scores of basic needs support. Although this study provides initial evidence that high programme quality can positively influence basic needs support, research is needed to further explore this in the literature. In their meta-analysis of after-school programmes, Durlak and colleagues (2010) called for identification of programme characteristics to help understand why some programmes are more successful than others and for clarification on how different aspects of programme quality influence different youth outcomes. The initial purpose of this study was to examine different aspects of programme quality and basic needs support from a community-based leadership programme; however, throughout the process of this research study, the initial leader of the programme left and a new leader was brought into the programme. In subsequent observations of the second leader, it was noted that the two leaders interacted with the youth in different ways. Little, Wimer and Weiss (2008) argued the quality of programme staff are a critical feature of high quality afterschool programmes; however, little research has explored how the quality of leaders may influence overall programme quality and basic needs support of youth. From observing the differences between the two leaders, the purpose of this study evolved to examine whether there were differences between the two leaders with regards to: (a) researcher observed programme quality and (b) youth perceived basic needs support. Given that this study was exploratory in nature no hypotheses were proposed.

Methods
Based on Petitpas et al. ' s (2005) recommendation for evaluating youth programmes, a case study approach was employed using a mixed-methods approach. Specifically, an embedded design was utilized, such that qualitative data played a supportive role to a larger quantitative study. The quantitative data were intended to gain an understanding of programme quality from the researchers' perspective and perceived basic needs support from youth's perspective. The qualitative data (semi-structured interviews) were used to provide depth of understanding to youth's experiences within the programme under both programme leaders.

Context and participants
This study was part of a larger study examining programme quality in youth programming and involved a leadership programme that was implemented for youth from low-income families between the ages of 13-18. The programme ran for 8 months from September to May and ran out of three clubhouses once per week in an effort to maximize the potential to reach youth. A total of 20 2-h sessions were delivered at each location over the course of the programme. The programme worked to integrate individual (e.g. personal choices and goal setting), family (e.g. supporting roles in the family), school (e.g. effort and achievement) and community (e.g. involvement and contribution). The main goals of the programme were to strengthen character and develop leadership skills. Youth were recruited into the programme if they were deemed at-risk, identified as having leadership potential, and interested in making a commitment to the programme.
Twenty-eight youth (9 boys, 19 girls) out of 41 youth involved in the programme obtained parental consent and completed the quantitative measures at two time points (mid programme and end of programme). These youth ranged between 14 and 18 years old (M age = 15.84, SD = 1.10). Youth identified their ethnicity as follows: 46.2% Caucasian, 20.5% Black/African American, 15.4% Asian, 7.7% multi-racial, 2.6% Aboriginal, 2.6% Arabic and 5.2% did not specify their ethnicity. One-quarter of youth were participating in the programme for the first time, while 36% for the second time, 18% for the third time, 13% for the fourth time, 3% for the fifth time and 5% did not indicate how many years they had been involved in the programme. Out of the 41 youth involved in the programme, nine dropped out of the programme during the second half and did not complete the measure at the second time point. Moreover, 15 (4 boys, 11 girls) youth engaged in a semi-structured interview that took place at the end of programming.
A maximum variation purposeful sampling technique (Patton, 2002) was used to recruit youth. This sampling technique allows for the gathering of youth perspectives of different gender, age, years of programme involvement and a representative sample from each of the three programme locations. Although more girls were interviewed, this was reflective of the programme participation make-up. All three researchers agreed that saturation had been reached after the 15th youth had been interviewed.
Two programme leaders were involved in programme delivery over the course of one year. Leader 1 was a 26-year-old female, of East-Indian descent, and was responsible for implementing the programme from September to December and withdrew from the programme due to unforeseen circumstances. As such, Leader 2, a 30-year-old Caucasian female was responsible for implementing the programme from January to April.

Procedure
Following ethical approval from the affiliated institution's Office of Research Ethics and Integrity, the lead researcher contacted the programme director and leader. From this, dates and times were arranged where the researcher attended the programme, introduced herself to participants, and explained what study involvement would entail. Parental consent forms were then distributed, completed and returned to the researcher. Over the course of the programme, two researchers observed 20 2-h sessions (11 facilitated by Leader 1, 9 facilitated by Leader 2). The 20 observations were spread across all three programme locations (7 observations each at two programme locations, 6 observations at the third location). Researchers took detailed field notes that helped supplement the completion of the Youth Programme Quality Assessment (YPQA; described below). Immediately after observing the programme session, researchers completed the YPQA.
One paper questionnaire was distributed to youth at two times points during the programme (halfway and at the end of the programme). This was initially done to examine if there were any changes in basic needs support over the course of the programme; however, coincidentally the week youth completed the first round of questionnaires midway through the programme, Leader 1 announced she was withdrawing from the programme. Therefore, while still examining changes in needs support in youth throughout the programme, the leader primarily responsible for facilitating needs support within the programme changed from time one to time two (time 1 = Leader 1, time 2 = Leader 2). The lead researcher distributed questionnaires to youth at the three programme locations which provided opportunities for youth to ask questions if needed.
As mentioned, interviews were conducted with a sample of youth. Semi-structured interviews took place during the last week of the programme either before, during, or after the session. All interviews occurred in person and were audio recorded. Prior to interviewing, youth were reminded of the voluntary nature of the study, their rights to confidentiality, there were no right or wrong answers, and what they discussed would not affect their programme involvement. Interviews were recorded using a digital audio recorder and lasted from 15 to 48 min (M = 30:31). Yohalem and Wilson-Ahlstrom (2010) argued that programme quality is best measured using multiple measures from multiple sources over multiple time points within a programme. Moreover, Hodge et al. (2013) argued that qualitative research is needed to comprehensively examine the important role that needs support plays within youth programming. Therefore, two perspectives (youth and external researchers) and three sources of data were attained (observations of programme quality using the YPQA, self-reports of basic needs using the Learning Climate Questionnaire and qualitative semistructured interviews to understand youth's experience in the programme) over two time points for the purposes of this study.

Youth programme quality assessment
Observations were conducted to assess programme quality and consisted of utilizing the YPQA to assess the contextual features of the programme. The YPQA is an observational tool used when conducting process evaluations within community programmes (High/Scope Educational Research Foundation, 2005). Flett, Gould, and Lauer (2012) have argued that there is an over-reliance on self-report measures, highlighting the need for more direct observational data collection; therefore, utilizing this tool can address this gap. The YPQA is used for programme improvement, monitoring and research and evaluation has been found to be a valid and reliable tool in both school and community-based programme settings for grades 4-12 (Yohalem, Wilson-Ahlstrom, Fischer, & Shinn, 2009). The YPQA is based off of the NRCIM's eight contextual features within programmes that are likely to promote positive developmental outcomes (Eccles & Gootman, 2002). The observational tool examines the context's safe environment, supportive environment, interaction and engagement (High/Scope Educational Research Foundation, 2005). Subscales fall under each domain, with each subscale having multiple items. There are 18 subscales and 63 items within the four domains. For each item, concrete descriptions are provided, and are rated on a scale from 1 (no evidence) to 5 (consistent evidence). It has been encouraged to use this tool at multiple time points throughout a programme, allowing for a thorough understanding quality over the course of a programme's entirety, as done in this study. Space was provided alongside each item to write down supporting evidence, which allowed for detailed accounts for a subsequent item scored.

Learning climate questionnaire (LCQ)
The LCQ was adapted from the Health-Care Climate Questionnaire (Standage, Duda, & Ntoumanis, 2005) and measures how well programme leaders meet the three psychological needs (autonomy, competence and relatedness) of youth participants. The LCQ is a 24-item measure that measures youth's perceptions regarding the degree to which their programme leader supported their sense of autonomy (15 items; 'I felt that the leaders provided us with choices and options'), competence (4 items; 'The leaders helped us to improve') and relatedness (5 items; 'I felt that the leaders were friendly towards us') and is measured on a 6-point scale from 1 (strongly disagree) to 6 (strongly agree). The LCQ has been was validated with adolescents in research examining needs support in a physical education setting and has good internal consistency . With the current sample, all three factors showed good internal consistency, with Cronbach's alphas ranging from .87 to .93.

Semi-structured interviews
An interview guide was developed for youth based on extensive review of programme quality and needs support literature. The interview guide aimed to explore youths' experiences of programme participation (e.g. 'Can you tell me a little bit about your involvement in [name of program]?') and their perceptions of basic needs support within the programme. Questions related to autonomy (e.g. 'Did you feel the leaders provided you with choices and options throughout the program? In what ways?'), competence (e.g. 'What did you learn in the program?'; 'How did the leader help you to develop [skill]?') and relatedness (e.g. 'Did you develop positive relationships with the leaders? In what ways?') were asked. Once the change of leaders became evidently pertinent to programme delivery, questions were added such as 'What was your experience like working with the leaders?'; 'I know there was a change in staff over the course of the program, how did you find that influenced you?' . Probes were used to explore participants' experiences further (e.g. 'Can you give me an example of what you mean by that?').

Control variables
Demographics were gathered from all youth in the programme. In the analysis of needs support, gender and ethnicity were controlled for based on previous research indicating the important effects of demographics on examined outcomes (Larson, 2000). As the programme was run at three locations, this was also used as a covariate to control for any differences across the three programme locations. All three variables were specified as categorical.

Data analysis
Quantitative data were analysed using SPSS 23.0. Descriptive statistics were calculated for the YPQA's four subscales and LCQ's three subscales. An inter-rater reliability analysis was performed using the Kappa statistic to determine consistency among researchers for the YPQA. To examine if there were significant differences between leaders related to programme quality, a multivariate analysis of variance (MANOVA) test was conducted on the four domains of programme quality outlined by the YPQA. Moreover, several analysis of variance (ANOVA) tests were run post hoc to further examine if there were differences between individual subscales within these four domains. Next, to examine if there were significant differences of youth perceptions of basic needs support between the two leaders, a repeated measures multivariate analysis of covariance (MANCOVA) was conducted. Effect sizes were calculated and reported for the domains of the YPQA and subscales of the LCQ using eta-squared (η 2 ) for the MANOVA's and Cohen's (1988) d for the post hoc analyses.
An inductive-deductive thematic analysis (Braun & Clarke, 2006) was used when analysing the qualitative data (interviews). This procedure allowed for the analysis to be driven by theory, understanding important topics within the literature deductively (e.g. Basic Needs Theory), while also including important elements that emerged inductively from participants' responses (e.g. differences between programme leaders). Specifically, a six-step procedure was utilized (Braun & Clarke, 2006). First, the researcher familiarized herself with the data by reading and re-reading transcripts and taking initial notes. Second, initial codes were generated in order to systematically identify and organize the data. Third, the codes were collated into potential themes. Next, the potential themes were reviewed to ensure codes fit in both their original context and the generated theme. Within this step, investigator triangulation (Yin, 2009) was utilized as a second and third researcher analysed the data. Discrepancies between the three researchers (e.g. labelling of themes, placement of quotes under themes) were discussed until agreement was reached. These changes occurred in the fifth step, where clear and concise names were provided to each theme that reflected the message in which the analysis was attempting to deliver. Finally, the most appropriate data extracts related to the research questions were selected that best illustrated the identified themes in relation to literature reviewed (Braun & Clarke, 2006). Pseudonyms were created for each quotation to protect the confidentiality of youth and leaders' names were replaced by 'Leader 1' and 'Leader 2' based on the order in which they facilitated the programme.

Results
Given this research used an embedded mixed-methods approach, where the quantitative data took on a primary role, the quantitative findings are presented first followed by the qualitative findings from youth interviews. Table 1 outlines the descriptive statistics (means and standard deviations) of the researcher observed scores of the four domains of programme quality as measured by the YPQA and the corresponding subscales within each domain in the first half of the programme facilitated by Leader 1 and the second half of the programme facilitated by Leader 2. When calculating the Kappa score, all assumptions were met and the inter-rater reliability for the researchers was found to be Kappa = .620 (p < .001), 95% CI (.561, .679) indicating a substantial strength in agreement for the observations (Landis & Koch, 1977). Preliminary analyses were conducted to ensure that there were no violations of the assumptions of normality, linearity, homogeneity of variances and homogeneity of regression slopes. As mentioned, to examine whether there were differences on the four domains of programme quality across the two programme leaders, a MANOVA was conducted. Results revealed a statistically significant difference between programme leaders on the combined dependent variables, F(4, 15) = 5.903, p = .005; Pillai's trace = .612; η 2 = .61. The dependent variables were examined separately using a Bonferroni adjusted alpha level of .013, and the results revealed that there were statistically significant differences for two of the four domains of observed programme quality between the two leaders. Significant differences were found within safe environment (F(1, 18) = 7.46, p = .014) and supportive environment (F(1, 18) = 11.77, p = .003), with Leader 1 having higher scores than Leader 2 on these two domains of programme quality. Both domains were found to have large effect sizes (1.25 and 1.56, respectively). There were no significant differences for scores across Interaction or Engagement.

Quantitative results
It should be noted that differences were examined across the subscales of the YPQA as well (see Table 1). Safe Environment measured both the physical and psychological safety of the programme environment, and upon further investigation of these subscales, emotional safety was scored significantly higher for Leader 1 and Leader 2 (F(1, 18) = 15.70, p = .001), with a large effect size of 1.73. Similarly, within Supportive Environment, the Warm Welcome and Reframing Conflict subscales were found to be significantly different, in that Leader 1 scored significantly higher than Leader 2 on both subscales (Warm Welcome: F(1, 18) = 15.65, p = .001; Reframing Conflict: F(1, 11) = 8.25, p = .017). It is important to note that not all observed sessions had incidents that required conflict management (n = 12; Leader 1 = 7, Leader 2 = 5); therefore, only these programme sessions were used in the analysis for this subscale. Both components were found to have large effect sizes (1.81 and 1.67, respectively). No other significant differences were found across the YPQA subscales; however, it should be noted that the Encouragement subscale within the Supportive Environment domain approached significance F(1, 18) = 3.888, p = .064 with Leader 1 having a higher mean than Leader 2 and this was found to have a large effect size of .89.
A repeated measures MANCOVA was performed to examine differences between youth's perceptions of needs support from the two leaders. Results revealed a statistically significant difference between leaders on the combined dependent variables of autonomy, competence and relatedness, F(3, 12) = 8.268, p = .003; Pillai's trace = .674; η 2 = .67. The dependent variables were examined separately using a Bonferroni adjusted alpha level of .017. Results revealed significant differences within the three subscales between time 1 and time 2 (see Table 2), whereby basic needs scores were perceived as being higher when the programme was facilitated by Leader 1 compared to Leader 2. Large effect sizes were also found for the three subscales ranging from .56 to .58. It is important to note that no significant differences were found when examining basic needs support across gender, ethnicity or between the three locations in which the programme was delivered.

Qualitative results
Qualitative results are presented in two sections outlining youth perceptions of how the two leaders: a) fostered or inhibited the sense of a psychologically safe environment, and as such, b) provided support, or lack thereof, for youth's basic needs of autonomy, competence and relatedness. It is important to note that based on youth's perceptions outlined within the quotations, overlap does exist between these two themes and when this occurred, is outlined within the context provided for each quotation.

Psychologically safe environment
Firstly, Shawna, who was participating in the programme for the first time, discussed how Leader 1 created a positive environment for the youth: 'The first leader [Leader 1], she had a more positive tone towards us and she was always uplifting' . Hannah, who had been involved in the programme for four years, reinforced that Leader 1 provided a psychologically safe environment where youth felt comfortable sharing within the programme and how that differed with the second leader: She came in thinking we were supposed to respect her and treat her like royalty. But there she is on her phone during session, but she tells us not to be on our phone or sometimes she just sits there eating when she could have been walking around helping answer questions or just chatting with us. So I feel like she wants us to respect her so much, but gives us so little.
Moreover, when asked how the leaders were able to create a psychologically safe environment, two youth discussed a strategy that Leader 1 utilized that they perceived as being successful in not only creating this environment, but also fostering relationships with youth (relatedness), as Sophia stated: Every couple of months [Leader 1] would check in on you. She was the kind of person that you could tell anything to and know she wouldn't tell … a very trusting relationship … it brings you closer and then you can understand them better, She knows a lot about me and we used to talk about stuff; we used to have one-to-one interviews and they were great' . In contrast, Michael noted that since the inception of the second leader, the one-onone meetings no longer took place, yet could prove to be beneficial in fostering relatedness between the leader and youth: 'We don't do those meetings anymore, which would be good to do though, it would help a lot. It would also help get her [Leader 2] closer to the members' .

Basic needs support
When asked what her experience was like working with the programme leaders, Lauren recalled a specific situation in which Leader 1 helped her: Michael perceived Leader 1 to provide a psychologically safe and supportive environment through the use of active listening, being non-judgemental and working together with youth to arrive at a solution; fostering a sense of autonomy and competence. In contrast, Leader 2 was seen as dismissing conflicts and often experienced resistance from youth because of lack of perceived mutual respect. Sierra seconded this notion through outlining the importance of challenging appropriately and allowing for creativity, which fostered the needs of autonomy and competence, or hindered them when not provided within a support environment: I feel [Leader 1] definitely challenged us and definitely didn't put the challenge in a stressful way. She put it as you had to use your brain to think and you had to put the puzzle together to figure out the solution and [Leader 2] challenges us, but sometimes I feel stressed out because it seems like there's only one right answer.
She went on to say: Similarly, when asked if they felt the leaders provided them with choices and options throughout the programme, two youth spoke of their disappointment in feeling as though they did not have a voice or choice when Leader 2 facilitated the programme. Hannah stated: Further, Shawna noted: 'A couple of times I've been a little angry at [Leader 2] because if we wanted to do something, there would be kind of a rebellious conflict or argument between what we wanted to do and what she expected us to do' . Both girls' quotations highlighted the lack of autonomy they felt within the second half of the programme. Lastly Bennett and Maya discussed how they were able to develop a sense of competence within the programme when facilitated by Leader 1 through the use of intentionally teaching life skills and corresponding strategies such as modelling appropriate behaviour and integrating a debrief at the end of programme sessions. Both youth did not perceive Leader 2 to effectively carry out similar strategies, which inhibited their understanding of programme lessons and components (e.g. competence). For example, Bennett stated: Honestly, I think when [Leader 1] was here it was a lot better. When she was a leader she was a lot wiser, so we learned a lot more -or I did. All the communication; she really went into describing why we needed those skills. It was just better and she was a better role model to look up to rather than [Leader 2].
Additionally, Maya discussed how the use of debriefs was seen as a useful strategy utilized by Leader 1 to reinforce the life skills learned within the programme and foster a sense of competence: [Leader 1] would always leave us something to think about at the end. I think it's better because then you actually you know what and how to use these skills we're learning in [name of program]. With [Leader 1], I would actually think about what she was saying, but now with [Leader 2], she often says we can leave five minutes early, and it's like 'what was the point of this session?' , there was no summary or take home, you know.

Discussion
The purpose of this study was to examine whether there were differences between two leaders in one youth leadership programme as it relates to researcher-observed programme quality and youth-perceived needs support. The event that occurred halfway through the programme, in which there was a change of leaders, allowed such findings to emerge completely inductively within the programme. Findings from this study revealed that Leader 1 was rated higher than Leader 2 in fostering both a safe and supportive environment. After further investigation, subscales of emotional safety, warm welcome and reframing conflict were rated significantly higher for Leader 1 than Leader 2, and encouragement approached significance. In regards to youth perceived needs support, significant differences were found across all three needs from time 1 to time 2, whereby youth perceived Leader 1 as being more effective in fostering their needs for autonomy, competence and relatedness than Leader 2. This was supported by qualitative findings as youth vocalized differences between leaders in terms of providing a psychologically safe environment and the support for these three needs.
As there were no other changes within programme structure or delivery other than the change in leadership from time 1 to time 2, it can be inferred that reported differences were due to this change. Moreover, although it has been argued that basic needs support can be attained from both adults (leaders) and peers (Deci & Ryan, 2000, 2002, the LCQ only measures perceived needs support by programme leaders, providing further justification for the changes in needs support being attributed to programme delivery and emotional climates created by the two leaders. As outlined, youth are more likely to thrive when in a social environment that supports their need for autonomy, competence, relatedness and a programme leader can play a role in either supporting or hindering these needs (Armour & Sandford, 2013;Bean & Forneris, (under review); Deci & Ryan, 2000) The quality of a programme's staff is one of the most critical features of a high quality programme (Eccles & Gootman, 2002;Little et al., 2008;Petitpas et al., 2005). One main strategy that Lerner (2004) outlined as of critical importance within youth programming is having positive and sustained adultyouth relations. Within this programme, the change in leaders may have hindered the fostering of basic needs support because the positive adult relationship that youth were experiencing was disrupted and not sustained over time. A study by Hetland and colleagues (2011) examined the relationship between leadership style and basic psychological needs fulfilment in the workplace. Researchers found that transformational leaders, recognized as role models who instil motivation and are intellectually challenging, were positively related to basic psychological need fulfilment. In contrast, transactional leaders, known for their controlling nature and results-based motivation practices, were found to be negatively associated with basic needs fulfilment (Hetland, Hetland, Schou Andreassen, Pallesen, & Notelaers, 2011). These findings align with the current study, based on the qualitative data, as Leader 1 was perceived by youth as a transformational leader (instilling competence and motivating youth to come to their own conclusions) and Leader 2 as a transactional leader (tolerating less youth input and taking less time to get to know youth one-on-one).
Previous research purports that youth's experience of trust in caring relationships with programme leaders is a critical contributor to programme outcomes (Strobel, Kirshner, O'Donoghue, & Wallin McLaughlin, 2008;Vandell, Larson, Mahoney, & Watts, 2015). Semi-structured interviews in this study mirrored results from quantitative analysis, and provided youth accounts of how Leader 1 was able to facilitate a higher quality programme related to psychological safety. Common themes that emerged pertained to issues of trust and respect between youth and leaders. Leader 1 was able to create a safe environment where youth felt both understood and respected (emotional safety), while Leader 2 was often regarded as judgemental. Leader 1 was reported as developing youth's trust by taking time for structured one-on-one's, while it was purported that no such time was set aside by Leader 2 to develop this rapport with youth. From this study, it can be inferred that certain characteristics of leaders, such as being trustworthy, respectful and interested in youth beyond regular programme interaction are thought to facilitate a high quality programme and accompanied needs support from youth. Similar findings were outlined in a study conducted by Bean and colleagues (2015) where the establishment of a trusting and caring environment and receiving positive support from the programme leaders in a physical activity-based life skills programme were critical components that helped to foster psychosocial development. Moreover, in a study conducted by Vandell and colleagues (2005) that examined two after-school programmes (one high quality and one low quality) for low-income youth, researchers found that the quality of staff-child relationships were critical to the facilitation of programme quality. The programme with a higher quality leader was successful in providing supportive interactions and also facilitating positive behaviour management, which related to the ability to meet youth's needs. Similarly, when leaders understood individual situations and circumstances of youth, they were better able to provide emotional support to them. This was similar in the case of the current study, particularly with the use of one-on-one meetings facilitated by Leader 1, outlined by youth as a critical strategy in fostering psychological safety and a sense of relatedness. Moreover, it can be inferred that Leader 1 was successful in delivering a high quality programme which allowed for the support of youths' basic needs.
It should also be noted that all subscales within the YPQA that emerged as being or approaching significant between the leaders were linked to providing a psychologically safe environment (e.g. emotional safety, warm welcome reframing conflict and encouragement). This finding can be discussed in relation to Maslow's (1943Maslow's ( , 1954 Hierarchy of Needs, the theory in which the YPQA was built upon. Within the hierarchy, each need must be satisfied before progressing to the next need. As such, safe and supportive environment (physiological, safety, love and belongingness) serve as a foundation for higher order needs of Interaction and Engagement (esteem and self-actualization; Flett et al., 2012;Maslow, 1954). Relating back to previously proposed strategies for fostering high quality programmes, it is evident that the primary elements proposed relate to the criticalness of providing physical and psychological safety within youth programmes. For example, Petitpas et al. (2005) first outline the importance of appropriate structured context and external assets, while Eccles and Gootman (2002) purport that physical and psychological safety, apprpriate structure and supportive relationships are fundamental. In the current study, Leader 2 did not successfully foster the foundational components, or what is referred to as the lower components of the pyramid; therefore, higher order elements, such as opportunities for choice and leadership, although provided by Leader 2, did not appear to be internalized by youth. This finding speaks to the criticalness of supporting psychological needs prior to working towards high order elements of programme quality. Findings from this study indicate a reverse of a previous proposition in using the YPQA, where it has been suggested that specifically targeting Interaction and Engagement (arguably higher order elements) can positive influence youth outcomes (Yohalem et al., 2009). While there is evidence to support this, findings speak to the importance of first fostering the foundational programme quality elements (safe and supportive environment) and once those are satisfied, working towards facilitating interaction and engagement in youth programmes.
As Akiva (2005) argued, outstanding programmes are those that foster all four domains of programme quality (safe environment, supportive environment, interaction and engagement). In this study, both leaders scored highly on interaction and engagement, suggesting they both facilitated a programme that encompassed key programme quality elements including opportunities for leadership, planning, choice and reflection. However, the results also illustrate that a programme can have high levels of interaction and engagement, without providing a safe or supportive environment. Without this safety and support, it appears that basic needs may not be supported and therefore youth are not able to have positive psychological growth experiences and hence reap all of the benefits from a programme that provides opportunities for interaction and engagement. Therefore, it is critical that researchers and practitioners ensure programme leaders are fostering all four domains of programme quality.

Strengths, limitations and future directions
This study had many strengths including the inductive approach that was taken to examine programme quality and needs support using a mixed-methods approach. Moreover, the same youth were involved in the same programme (context); therefore, the only different variable was the leader responsible for programme delivery. This change in leadership allowed for a similar environment that would be constructed within an experimental design within a controlled environment, yet occurred inductively. Study limitations must also be recognized. First, although data collected were from youth who consistently participated in the programme, data were based on one particular programme; therefore, generalizability of results is limited. Moreover, the lack of a control group limited researchers' ability to isolate the effects of the programme and generalize results beyond the study population. Second, often the case in programme evaluation, there is potential of social desirability with self-report measures and semi-structured interviews. However, the researchers tried to limit this by reminding youth that there were no right or wrong answers and that it was important to be honest. Moreover, researchers used observational measures and provided objective evidence that supported youth perceptions. Lastly, while Ryan (2000, 2002) identify that basic needs can be supported by various individuals (e.g. via peers and adults), the LCQ only measures youth's need support with reference to a leader-fostered climate. While useful within this study, it is recognized that future research is needed that examines the influence of other important individuals as it relates to needs support within a youth leadership programme context.
As this is one of the first studies to study the importance of programme quality features within a leadership programme (e.g. psychologically safe environment), future research needs to investigate the relative importance of contextual features to ultimately foster high quality youth programming that supports youth's basic needs. Moreover, as the supporting of basic needs has been argued to foster psychosocial outcomes, future research should utilize a measure of youth developmental outcomes to understand if and how programme quality and needs support relate to these outcomes. Moreover, despite the surmounting evidence that it is critical to foster psychological safety, the YPQA combines physical and psychological safety into one programme setting feature. Future research or tool development related to programme quality may involve dividing the currently combined elements as, although the constructs are related, they measure very different components of safety and thus should be measured separately, as without rigorously examining individual items and subscales, there is the potential to misinterpret or skew the overall rating of programme quality.

Conclusion
The present study has significantly contributed to the existing body of literature by examining how programme leaders may directly influence the overall quality and psychological needs support of youth participants. Although a number of researchers have outlined what a high quality programme consists of, few studies have begun to examine the role that leaders play in delivering high quality programmes and the influence this has on youth's basic needs support. Moving forward, findings from this study speak to the criticalness of programme delivery and subsequent training on behalf of programme leaders, and may help explain why some programmes are more successful than others, despite being well intentioned and sharing many common elements. In sum, with as many as 86% of youth in Canada participating in at least one organized extra-curricular (Guèvremont et al., 2014), such as this out-ofschool leadership programme, it is essential that programmes are implemented the right way, and by the right leaders, so that youth can reap the full benefit of participation.

Notes on contributors
Corliss Bean is a PhD candidate at the University of Ottawa. Her research focuses within the domain of positive youth development, and specifically, the development, implementation and evaluation of youth programmes. Examining the influence of programme quality on basic needs support and psychosocial outcomes is at the core of her study interests. Recently she has been working with various community youth organizations to aid in the facilitation of high-quality programmes that enable opportunities for youth to develop the necessary skills to deal with the demands of everyday life.
Meghan Harlow is in doctoral degree at York University. Meghan's research interests broadly encompass long-term youth development in sport, as well as exploring sport participation amongst marginalized populations. With both applied and theoretical experience in the field, Meghan hopes to apply a unique lens and experience in sport to her own research in the area.
Tanya Forneris, PhD Counselling Psychology, is an associate professor in the School of Human Kinetics at the University of Ottawa. She has a background in sport and counselling psychology. Her research interests lie in the domain of positive youth development, and more specifically, the development, implementation and evaluation of life skills based (e.g. goal setting, confidence building, leadership) sport and physical activity programmes to enhance youth development, particularly with marginalized populations.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Social Sciences and Humanities Research Council of Canada.