Reference no: EM132344225
Part A
Questions to be answered
1. A skill acquisition researcher is interested in examining the effects of practice variability on learning of golf skills. She aims to set up an experiment in her laboratory where she will examine the effects of different practice schedules on a simulated putting skills test. She proposes to have a total of 60 participants who will be randomly assigned into one of two practice conditions. Half of the participants (n = 30) will complete the practice trials according to a constant practice schedule where they will complete the 40 practice trials of the putting skill from adistance of 3 metres. The other half of the participants(n = 30) will complete the practice trials according to a variablepractice schedulewhere they complete 40 practice trials of the putting skill from four different distances (10 trials at each of 1, 3, 5 and 7metre distances) in a random order. Performance will be measured using a putting skill test (10 putts) over a distance of 4 metres before and after participants complete the 40 practice trails. The researcher aims to ensure that conditions (e.g., temperature, lighting, noise) during the experiment will be carefully controlled. In discussing her proposed experiment, a colleague of the researcher questions the design-specifically he questions the external validity of the experiment. The researcher is prompted to reflect on the internal and external validity of her proposed experimental design. What is meant by the terms ‘internal validity' and ‘external validity' (briefly define each term) and explain how each affects the quality of an experimental study.
2. A researcher is interested in the relationship between amount of weekly physical activity and mental wellbeing among university students. She collects self-report survey data from 500 undergraduate students. The survey is used to capture data on a number of measures including: time engaged in moderate-vigorous physical (MVPA) activity over the past two weeks; and general wellbeing measured using the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS). The WEMWBS consists of 14 positively worded items relating to experiences over the past two weeks (e.g., ‘I've been feeling relaxed') and responses to items are indicated using a 5-point Likert scale (1= ‘none of the time'-5= ‘all of the time'). Scores on the WEMWBS range from 14-70 with higher scores indicating better individual wellbeing. After collecting, inputting and checking her data (all found to be okay), the researcher conducts a (Pearson) correlational analysis to examine the relationship between the two variables and obtains the following results.
Correlations
|
|
MVPA (hrs/2weeks)
|
Mental wellbeing score
|
MVPA (hrs/2weeks)
|
Pearson Correlation
|
1
|
.420
|
Sig. (2-tailed)
|
|
.041
|
N
|
500
|
500
|
Mental wellbeing score
|
Pearson Correlation
|
.420
|
1
|
Sig. (2-tailed)
|
.041
|
|
N
|
500
|
500
|
Help the researcher interpret her findings by answering the following questions (make sure you number your responses to each of the separate questions):
i. What is the obtained correlation value between MVPA and mental wellbeing?
ii. Is the correlation significant (using α= .05)?
iii. What is the direction of the correlation?
iv. What is the coefficient of determination for the obtained correlation?
v. What does the coefficient of determination value indicate about the relationship between MVPA and mental wellbeing?
3. What are the differences between quantitative and qualitative research methods in terms of the forms of information (data) that are gathered? Provide examples of common methods used to collect the different types of data in exercise and sport science.
4. Different methods can be used to acquire knowledge and solve problems. Broadly these can be distinguished as unscientific or scientific methods. Describe two unscientific methods of acquiring knowledge/problem solving and then describe the scientific method and explain how/why the scientific method is useful for thinking about and solving problems.
5. A researcher is about to conduct an experiment to examine whether post-exercise consumption of a supplement containing whey protein, amino acids, creatine and carbohydrate combined with a strength training program promotes greater gains in muscle strength compared with an isocaloric, carbohydrate only control combined with strength training. The researcher proposes to randomise a total of 30 participants into two groups (n=15 per group). Previous studies focusing on similar supplements in combination with weight training reveal an effect size of 0.6. He is proposing to analyse his data (using a statistical test known as a t-test) with alpha set at .01 but is concerned that given his study design (i.e., participant numbers and ‘known' effect size), his analysis may have insufficient power. Explain what is meant by the term power and then describe two actions the researcher could take to increase the power of his study and note any advantages/disadvantages associated with these actions?
Part B
SPSS analyses and reporting and interpreting results
For the following questions you will need to download the datafile ‘HSE104 AT1 Tennis.sav' from the Assignment 1 folder on Moodle. You should download and save a copy of this file in your home directory so you can navigate to find this file.
The extracted ‘tennis' datafile is a freely accessible data set (hence it is not de-identified) that has been modified and includes summary match statistics for a set of elite-level tennis matches. The datafile contains 11 variables and these are described as follows:
• ‘winner' - Name of the winning player
• ‘loser' - Name of the losing player
• ‘gender' - Gender of players [1=female; 2 = male]
• ‘matchtime' - Time of day match was played [1 = day; 2 = evening/night]
• ‘surface' - Type of court surface [1= synthetic; 2 = clay; 3 = grass; 4 = cement]
• ‘ral_len' - Average rally length
• ‘shot_per_sec' - Shots per second
• ‘pc_first' - Percentage of first serves in
• ‘first_pc' - Percentage of points won on first serve
• ‘second_pc' - Percentage of points won on second serve
• ‘ln_ral_len' - Natural log of average rally length (i.e., this variable has been created by "transforming" data from the original variable ‘ral_len')
You should use a University PC to access the SPSS program to answer these questions.(NB. If you choose to use SPSS via Apps on Demand [i.e., browser-based SPSS] then you should be clear about file handling processes and allow plenty of time to perform the analyses as well as output file export).
Using SPSS you should open the ‘HSE104 AT1 Tennis.sav' datafile (File menu → Open → Data) and then familiarise yourself with the datafile, variables, variable labels...and actual data-points. You should then perform analyses to answer the following questions.
1. Prior to commencing any analysis it is important to carefully inspect the contents of the datafile for missing data or out of range values or other anomalies. Checking the contents of your datafile precedes any ‘cleaning' work such as inserting or replacing values or adjusting the name of a variable, value labels or the ‘type' of variable.
You should use SPSS for this task. Within SPSS you should check the datafile and complete Table 1 below. Note you are not asked to alter any scores/variables (i.e., ‘clean the data') just simply check and report observations/results of your checks.
Table 1. Data checking results.
Check performed
|
Response
|
How many cases (or records) are there in the datafile?
|
|
How many 'valid' cases are there for the variable 'surface' (i.e., cases that are not missing data on this variable)?
|
|
Overall, how many cases have complete data (i.e., cases that are not missing any data on any of the 11 variables)?
|
|
How many matches were played on a clay surface?
|
|
Inspect the variable 'pc_first'. What was the highest percentage of first serves in recorded by a player?
|
|
What 'type' of variables are the 'winner' and 'loser' variables?
|
|
2. Prior to undertaking any analysis using parametric statistics, it is important for researchers to evaluate whether scores on a variable meet the assumption of normality: whether the variables of interest are normally distributed (or not). You are required to test and report on the distributions for two groups on the following variables:
• ral_len
• pc_first
• first_pc
You should use SPSS for this task. Once you have finished testing these variables for normality, refer to the SPSS output to complete Table 2 below.
Table 2. Normality testing results.
DV
(Variable)
|
IV
(Gender)
|
Kolmogorov-Smirnov test results
|
Interpretation (i.e. is the distribution normal or non-normal?)
|
|
Test statistic
|
p value
|
example1
|
Female
|
0.03
|
0.53
|
For females, scores on 'example1' were normally distributed.
|
Male
|
0.53
|
<0.01
|
For males, scores on 'example1' were not normally distributed.
|
ral_len
|
Female
|
|
|
|
Male
|
|
|
|
pc_first
|
Female
|
|
|
|
Male
|
|
|
|
first_pc
|
Female
|
|
|
|
Male
|
|
|
|
3. Another important check that researchers will perform prior to parametric statistical analysis is to test that scores on a variable of interest have approximately equal distributions (or equal variances) for two (or more) groups. You are required to test for equivalence of variances for the following variables for male versus female players:
• ral_len
• pc_first
• first_pc
You should use SPSS for this task. Once you have finished your analysis, refer to the results of your variance testing to complete Table 3 below.
Table 3. Homogeneity of variance test interpretations.
Variable
|
Levene statistic
|
p value
|
Interpretation
|
example1
|
1.36
|
0.59
|
Levene's test result (1.36) was not statistically significant (p = 0.59); variances for 'example1' for females and males were equal.
|
example2
|
8.76
|
0.01
|
Levene's test result (8.76) was statistically significant (p = 0.01); variances for 'example2' for females and males were not equal.
|
ral_len
|
|
|
|
pc_first
|
|
|
|
first_pc
|
|
|
|
4. Central tendency and variability values are typically calculated during exploratory data analysis, to condense the scores for each variable of interest into summary values that can help researchers develop an initial understanding about the structure of the data. This can be either done for the whole sample or for different subgroups. You are required to compute means and standard deviation for the following variables for females and males:
• ral_len
• pc_first
• first_pc
You should use SPSS for this task. Once you have finished your analysis, refer to your central tendency and variability results to complete Table 4 below.
Table 4. Central tendency and variability results.
DV
(Variable)
|
IV
(Gender)
|
Mean
|
Standard deviation
|
Brief interpretation
|
example1
|
Female
|
80.52
|
2.12
|
The mean for 'example1' was 80.52 (SD = 2.12) for females and 77.61 (SD = 2.07) for males.
|
Male
|
77.61
|
2.07
|
ral_len
|
|
|
|
|
|
|
|
pc_first
|
|
|
|
|
|
|
|
first_pc
|
|
|
|
|
IV = independent variable; DV = dependent variable
Presentation, referencing and SPSS output file
Presentation-Including spelling, grammar, clarity, and adherence to word limits/format requirements
Referencing-Accuracy of referencing (i.e., citing sources appropriately; correct reference list)
SPSS output file-Asingle PDF copy (NOT the ‘spv' SPSS output file) of your cleaned SPSS output (i.e., a file that contains only figures / tables / text that are relevant for the five questions; all other ‘extraneous' information can be deleted from your output file so that your PDF file size is manageable).
Attachment:- Research Methods and Statistics in Exercise and Sport.rar