Short Answer: Yes!

Dr. Ana Catarina Vieira de Castro has published multiple studies which compare the effects of aversive-based and reward-based dog training methods.

Let’s take a look at one of those studies, titled “Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare”.

Summary: Compared to dogs trained using aversive or mixed methods, dogs trained using reward-based methods were visibly more relaxed and produced less cortisol during training. Dogs trained using reward-based methods also needed less repetitions to learn a new behavior and displayed less latency in the finished behavior.

In this study, Dr. de Costa and five collaborators measured stress by counting the frequency of stress behaviors in video recordings and analyzing cortisol levels in saliva samples. Dog participants were separated into three groups: Group Reward, Group Aversive and Group Mixed.
Dogs in Group Aversive and Group Mixed had significantly higher numbers of stress-related behaviors than dogs in Group Reward. Group Aversive’s post-training cortisol levels were also higher than Group Reward.
In addition to measuring stress, the study also compared the efficiency of these three training methods. They found that in terms of repetitions needed to learn a new behavior, Group Mixed needed more repetitions than Group Reward. In terms of latency and the speed at which the behavior is performed, Group Aversive was slower than Group Reward. This evidence suggests that in addition to producing less stress, reward-based methods also produce faster learning overall.

I kept this as simple as possible, while still covering key findings. If you’re interested in science-based dog training and have some experience interpreting behavioral statistics, I highly recommend checking out this awesome study!
https://doi.org/10.1371/journal.pone.0225023