Assessment Scenario 1: DS1
The aim of this evaluation is to analyze a data set gathered over eight weeks’ time for a group-based physiotherapy program to assess whether there are any significant changes over this period regarding the participant’s Pain, Rating of Perceived Exertion. The null hypothesis for these tests was that there is no strong correlation between the variables mentioned above and weight which was tested using the VAS scale questionnaire. To test the hypothesis, some tests were conducted.
Descriptive Frequencies Tests for Demographics
Table 1 below shows the frequency result of the Gender data from the dataset. It can be seen that from a total of 50 respondents, 20 were male and 30 respondents were female.
Running the descriptive statistics tests for the Age in Years variable of the data set (Table 2), it can be seen that the minimum age of respondent is 19 whereas the maximum age of respondents is 40 years. The average age of the data set is 29.33 years.
For analyzing the weight in Kg of a dataset (T1), it can be seen from Table 3 below that the average weight of respondents is 89.59 Kg whereas 85 is the minimum and 95 is maximum.
For analyzing the weight in Kg of a dataset (T2), it can be seen from Table 4 below that the average weight of respondents is 78.82 Kg which is less than the average of T1, whereas 73 is the minimum and 82 is maximum.
A normality test was conducted for the variable Pain T1. It can be observed in the set of Table 5.2 gives the Std. Deviation 5.306 which shows that the data set is scattered and is not clustered around a particular reading.
Skewness, as per the results in Table 5.2. are positive thus indicating positive (right) skew; whereas kurtosis value is negative; a negative one indicates negative kurtosis (Mayers, 2013). Moreover, the Std. Error for both is less than ±1.96 suggesting that the departure from normality is not too extreme.
A normality test was conducted for the variable Pain T2. It can observe in the set of Table 6.2. Gives the Std. Deviation 3.876 which shows that the data set is scattered and is not clustered around a particular reading.
To understand the relationship between different variables, some tests were conducted. The first statistical test conducted was between variable RPE on a 0-100mm scale * Pain on a 0-100mm scale (Table 7). The Chi-Square test carried out on the data was significant at the 0.238 level (2-tailed p>0.05) of significance. (χ 2 = 2.450E3, df = 2401) so we conclude that there is a similarity in the representation of both variables thus proving the both variables are somewhat dependent, hence our null hypothesis is rejected.
The same results were obtained when the tests were run between T2 and RPET 2 variables. Table 8:
A paired Samples statistics tests were conducted between four variables in a pair of 2 (Table 9). Pair 1 calculated the correlation between weight t1 and pain t1 and pair 2 calculated the correlation between weightt2 and pain t2.
The paired sample test show a Sig (2-tailed value) of .000 for both pairs thus shows a significant relationship between variables. With a Std Dev of 5.769 for pair 1 and Std Dev of 4.936 for pair 2, so we conclude that there is a similarity in the representation of both variables thus proving the both variables are highly dependent. Thus, it can be said that weight is directly correlated to the pain, thus rejecting our null hypothesis.
The aim of this study is to establish if there are differences between the three groups having completed two different back pain interventions (group-based sessions and one-to-one physiotherapy sessions). Respondents from two different clinics were recruited, and some tests from that data were conducted. The data is categories into variables including Ratings of participant Pain and RPE; in addition to their ‘Intention to Continue’ with the intervention. Pain, RPE, and Intention to Continue scores were recorded using a visual analog scale (VAS; a psychometric response scale which can be used in questionnaires using a 100mm line). The null hypothesis for this assessment is that there is no difference in the feeling of pain for both invention groups.
Descriptive Frequencies Tests for Demographics
The second data set was analyzed using SPSS software. To first test of the analysis was a basic descriptive statistics of the variable Gender. As it can be seen that the total number of respondents were 30 with Std Dev. Of 0.509. Table 1 shows the statistics of the test:
Moving further, the frequencies test was run for the variable intervention that was represented by three internal variables, 1 for Group-based intervention, 2 for individually-tailored intervention and 3 for exercise leaflet. The three variables were used ten times each thus having the similar percentage of frequency as shown in Table 2 below:
The third demographic test was run for the variable Age (years). The minimum age of respondents was 41 whereas the maximum age was 49. The Std. Deviation of 2.245 (Table 3) shows that there is not much difference in a range of respondent’s age.
Normality tests were conducted between various variables to analyze the data patterns of the dataset. The variable intervention held high importance in this set as it focused on the intervention method used by the respondents for back-pain. The two most important readings in the table below are Skewness and kurtosis. Moreover, the Std. Error for both is less than ±1.96 (1.52) suggesting that the departure from normality is not too extreme (Table 4)
The steam and leave test conducted for the variable intervention to show the same results for all three sub-categories as the frequency of responses is the same i.e. ten each (n=30).
Intervention Stem-and-Leaf Plot
Frequency Stem & Leaf
10.00 1 . 0000000000
.00 1 .
10.00 2 . 0000000000
.00 2 .
10.00 3 . 0000000000
Stem width: 1
Each leaf: 1 case(s)
The next test conducted was T-Test to analyze any statistical significance between the two variables (Noru, 2012) Pain score and RPE. The two variables were paired together to the correlations test, thus showing a significance level of 0.001 (Table 5) which Sig 2 value of 0.004 thus concluding a strong dependence between both variables of pain and effort by the respondents., hence rejecting our null hypothesis.
It can be seen that there is a negative correlation value between the variable pain and RPE thus confirming relationship between two variables that if one variable increases other decreases (Ntoumanis, 2001), and vice versa, hence rejecting our null hypothesis.
The next test was that of Crosstabs to describe the interaction between two categorical variables, i.e. Pain Score and intention to continue with intervention.
The Chi-Square tests table below shows the different values of the relationship between these two variables. As the alpha value is .05 (Norusis, 2008), and our “Asymp. Sig. (2-sided)” for the Pearson Chi-Square statistic is greater than .05 (.252). Thus there is no a relationship between the variables based on the level of pain and use of intervention, hence supporting our null hypothesis.
Assessment 3 focused on analyzing participants ‘Intention to Continue’ with the program in a back pain clinic. Two variables hold great importance in this data set: Intention to Continue’ based on the participants ‘Effect Expectancy’ (the degree of ease associated with the program). And for that purpose, the data collected will be used to assess the appropriateness of using ‘Intention to continue’ and ‘Effect Expectancy’ as a prediction model and use these findings to report the likely ‘Intention to Continue’ score for a person with an ‘Effort Expectancy’ score of 50. And for that purpose, the null hypothesis for this assessment that there is no strong correlation between effort expectancy and intention to continue.
Descriptive Frequencies Tests for Demographics
The frequency analysis was conducted for the variable intervention that showed the data for respondents that were focusing on group self-management intervention. With total 40 respondents, the increasing percentage was calculated to be 100% thus ensuring there were no missed respondents (see Table 1).
Gender of participants was analyzed using the simple frequency test. Table 2 below shows that there were 20 male and 20 female respondents with a valid percentage of 50 each (n=40).
The next variable analyzed using SPSS detailed stats was Age. 46 was the minimum age of the respondent, and 61 was the maximum age with an average age of 54.73. The standard deviation of 3.388 shows not mildly significant variation between an age of respondents (see Table 3).
Paired Samples Test was conducted between four variables. Pair 1 focused on the effect of expectancy of the respondents in pain and intention to continue the intervention, and Pair 2 analyzed the response and planned to continue. It can be seen that the Sig of Pair 1 is 0.005 which is thus representing a high correlation between the two variables.
As for Pair 2, the significance value is 0.00. Our hypothesis here was that both variables do not have a relation. However, a low p-value (such as 0.00 see Table 4.2.) was took as evidence that the null hypothesis can be ‘rejected.’
Table 4.3 shows the Sig (2-tailed) value of Pair 1 and Pair 2. Both pairs have a 0.00 reading thus proving a significant relationship between Pair 1 variables and Pair 2 variables. Therefore the effect expectancy which provides data on the degree of ease associated with the program with the intention of using the program is significantly correlated. Also, intervention and plan to continue the program are also significantly correlated.
The statistical test of ANOVA (Analysis of Variance) was conducted for the variable ‘Effect Expectancy’ (the degree of ease associated with the program) for between groups and within groups. As the value of Sig. is 0.289 which is greater that the alpha value of 0.05, therefore, there are non-significant effects between the groups.
To analyze the data for intention to continue and effect expectancy value of 50 a statistical test was conducted. It can be seen from the table below that only case 34 with 50 effect expectancy opted for intention to continue. As the number of respondents with 50 effect likelihood is slight, therefore we can conclude any significance from the dataset.
|Case Number||Effect Expectancy|
|Intention to Continue||9||1||1||36|
|a. Limited to first 50 cases.|
Mayers, A., 2013. Introduction to Statistics and SPSS in Psychology. Pearson.
Norusis, M., 2008. SPSS 16.0 statistical procedures companion. Prentice Hall Press.
Noru, M.J., 2012. IBM℗ ʼ SPSS℗ ʼ Statistics 19 Guide to Data Analysis. Prentice Hall.
Ntoumanis, N., 2001. A step-by-step guide to SPSS for sport and exercise studies (pp. 146-149). London: Routledge.