The Effectiveness of a 6-Week Running Weight Loss ProgramReportII - Report Example

Independent sample t-test Brief of the study This study used the data file d “A study of the effectiveness of a 6-week running weight loss program”. The study involved 30 individuals who volunteered to participate in a 6-week weight loss running program. The researcher runner met at a local running specialty shoe store twice during the week (Monday to Friday) as a group over 6-week period and once on weekends at a local park (either Saturday or Sunday). Those who entered the program were interested in losing weight. A group plan was created to follow over the 6 week period by the researcher/trainer. The subjects were asked to log their mileage throughout the program. Research question(s) and hypotheses The primary purpose of this study was to analyze data collected from researchers estimating the effect of gender on pre self-esteem of weight loss program participants. The researchers tested null hypothesis that there is NO statistically significant difference in runners’ pre self-esteem score by gender against an alternative hypothesis of there is a statistically significant difference in runners’ pre self-esteem score by gender. The following is the primary research question: Does gender have an effect in pre self-esteem of weight loss participants? Variables The variables for this study were demographic information that included information on the participant’s sex and history or running (yes or no). The base line data was made of pre self-esteem variable. Gender which was the independent variable in the study was categorical in that one was either male or female. Pre self-esteem, the dependent variable was continuous where the research subjects could give any value as their pre self-esteem level. For the purpose of this report, the authors run the frequencies of the variables to check the characteristics of the data collected. See Table 1 for the frequencies of gender variable. The table indicates that there were 17 males and 13 females. Although there are not equivalent sample sizes by gender, the design is balanced according to Meyers, Gamst, and Guiarino (2006) criteria. The criteria states that, a design is balanced when the largest cell number is not 3 times greater than the smallest cell number. Table 1 Frequencies by Gender Frequency Percent Valid Percent Cumulative Percent Valid Male 17 56.7 56.7 56.7 Female 13 43.3 43.3 100.0 Total 30 100.0 100.0 Further exploration of the variables involved checking the effect of gender to Pre self-esteem variable. Table 2 gives the descriptive statistics of Pre self-esteem by gender. This shows that female scored highly on the pre self-esteem test than male. Table 2 Descriptive statistics of Pre self-esteem variable by gender Sex Statistic Std. Error PreSelfEsteem male Mean 29.24 1.538 95% Confidence Interval for Mean Lower Bound 25.98 Upper Bound 32.49 5% Trimmed Mean 29.54 Median 29.00 Variance 40.191 Std. Deviation 6.340 Minimum 16 Maximum 37 Range 21 Interquartile Range 12 Skewness -.470 .550 Kurtosis -.805 1.063 female Mean 30.46 1.580 95% Confidence Interval for Mean Lower Bound 27.02 Upper Bound 33.90 5% Trimmed Mean 30.46 Median 30.00 Variance 32.436 Std. Deviation 5.695 Minimum 22 Maximum 39 Range 17 Interquartile Range 9 Skewness .000 .616 Kurtosis -1.052 1.191 Data analysis methods Independent sample t-test was adopted for this study. The test was to check if there was a significant difference in pre self-esteem of runners by gender (Male and female). This statistical test is appropriate when one has separate samples under the same treatment condition (Heiman, 2013). In this case the treatment was running. The population mean and standard deviation are normally not known. The key assumptions for the test are the sample populations under study must be independent, the scores in each population must be normally distributed, and the two populations must have equal variance (Sheskin, 2003). From the descriptive statistics above, the difference in the means indicate that these samples were independent of each other. The researcher screened the data before the actual analysis by independent sample t-test. Frequency Table 1 above indicated that there were no missing values. The box plot also indicated that there were no outliers that could reflect data entry errors in the study. The skewness and kurtosis values were within the acceptable range of -1.0 to 1.0 (Meyers et al. (2006). Results An independent sample t-test of pre self-esteem by gender was conducted. Male (n=17, M=29.24) and female (n=13, M=30.46) had standard deviations SD=6.34 and SD=5.7 respectively. This indicates that the two were not further spread from their means since their standard deviations were within acceptable range. The means of the two genders indicated that female gender scored higher than male on the pre self-esteem scale (see Table 3 below). Table 3 Descriptive Statistics for Pre-self steem Score by Gender Gender n M (SE) SD Male 17 29.24 6.34 Female 13 30.46 5.60 Homogeneity of variance on the pre self-esteem data collected by gender was assessed using Levene’ test of equality of variance. Levene test indicated non-significant results, F=0.494, p>0.05. This indicated that the assumption on homogeneity of variance was met. The samples varied in testing for normal distribution as the pre self-esteem for male had a small negative skewness while pre self-esteem for female had zero skewness. According to Meyers, et al. (2006) criteria, skewness was within tolerance level of between -1.0 and 1.0 but kurtosis was not. The data was however adopted for further analysis. Male Skewness -.470 Kurtosis -.805 Female Skewness 0.000 Kurtosis -1.052 Independent sample t-test results showed that there was no statistically significant difference in average pre self-esteem score by sex, t 28=,-.548 p > .05. This means that the researchers failed to reject the null hypothesis and conclude that there is no statistically significant difference in runners’ pre self-esteem score by gender. According to Cohen (1992), the effect size of the study is medium since d=0.793. This is supported by the standard deviation units of genders. Male (SD=6.34) and female (SD=5.60) indicate that the values were not so much spread and thus could have a medium effect. References Cohen, J. (1992). Quantitative methods in psychology. Psychological Bulletin, 112(1), 155-159. Heiman, G. W. (2013). Basic Statistics for the behavioral sciences. Belmont CA: Cengage learning. Meyers, L. S., Gamst, G., & Guarino, A. J. (2006). Applied multivariate research: Design and interpretation. Thousand Oaks, CA: Sage. Sheskin, D. J. (2003). Handbook of parametric and nonparametric statistical procedures ed. Boca Raton, Fla: Chapman & Hall/CRC Press. Dependent samples t-test Brief description of the study This study used the data file named “A study of the effectiveness of a 6-week running weight loss program”. This study involved 30 individuals who volunteered to participate in a 6-week weight loss running program. The researcher runner met at a local running specialty shoe store twice during the week (Monday to Friday) as a group over 6-week period and once on weekends at a local park (either Saturday or Sunday). Those who entered the program were interested in losing weight. A group plan was created to follow over the 6 week period by the researcher/trainer. The subjects were asked to log their mileage throughout the program. Research question(s) and hypotheses The primary purpose of this study was to analyze data obtained from researchers investigating if there is a significant change in weight to runners after 6-week weight loss running program. The researchers tested the null hypothesis that there is NO statistically significant change in runner’s weight after 6-week weight loss running program against an alternative hypothesis that there is statistically significant change in runner’s weight after 6-week weight loss running program. The following is the primary research question: Does weight change after engaging in 6-week weight loss running program? Variables The dependent variable for this study was the weight scales and it was a continuous variable. These were pre-weight, and post weight. The research subjects could have weight of any value that is not subjected to any scale. In this study there were 27 (50%) PostWeight subjects and 27 (50%) PreWeight subjects. The independent variable that was continuous was the runners’ weight subjected to Pre running and Post running conditions of runners participating in the weight loss program. Pre Weight and Post Weight thus present a dichotomous nature that is recommended for an independent variable of a dependent sample t-test (Nimon, 2012). With this sample, the mean weight was 174.93 (SD=18.58) for PostWeight subjects and 179.81(SD =18.661) for PreWeight subjects as indicated in Table 1. This indicates that PreWeight had a significant higher mean in the study. Power analysis, that is used to calculate minimum sample size, estimate that the minimum sample size should be 20 give a power of 0.80 (Myoung, 2010). This study had 27 participants meaning it sufficiently satisfied power analysis estimate. Table 1 Paired Sample Statistics Tables (N = 27) M SD Pair 1 PostWeight 174.93 18.58 PreWeight 179.81 18.66 Data analysis methods A dependent sample t-test was conducted to test if there is a statistical significant change in runners weight after 6-week weight loss running program. This statistical test is appropriate for testing the difference in means for research designs that include two measurements taken on the same subject (Sheskin, 2003). This design was perfectly balanced since the PostWeight (27) subjects equaled the PreWeight(27) subjects. The researchers screened the data prior to the analysis and the results indicated that there were no missing values. Dependent sample t-test works on the assumptions that, there are no significant outliers in the difference between the two related groups, and the difference between the two groups is approximately normally distributed (Elliott & Woodward, 2007). The box plot helped in checking if there exist significant outliers by gender. There were no significant outliers for weight scores by gender. Test for normality for the dependent variables by gender was conducted by use of paired samples correlations and checking for skewness and kurtosis. In this study, skewness of -0.326 and kurtosis of -0.77 were both within the expected range of between -1.0 and 1.0 (Meyers et al. 2006). A higher correlation indicated that there was a statistically positive correlation between weight for pre weight and post weight subjects after the 6-week program. Results To test if there was statistically significant differences in weights between pre-weight and post-weigh scales taken before the start of the running program and after, a dependent sample t-test was used to test the null hypothesis that there is NO statistically significant change in runner’s weight after 6-week weight loss running program. There was a total of 27 individuals whose average weights were PostWeight (M=174.93) and PreWeight (M=179.81). Pre weight was significantly higher than post weight. The two sample populations shared a significantly higher correlation of r=0.992, p Accessed 25 November 2014. Nimon, K. F. (2012). Statistical assumptions of substantive analysis across the general linear model: A Mini-Review. Frontiers in Psychology, 3,322. Sheskin, D. J. (2003). Handbook of parametric and nonparametric statistical procedures ed. Boca Raton, Fla: Chapman & Hall/CRC Press. One Way-ANOVA Brief description of the study This study used the data file named “Sesame Street”. The data set is part of a large data set collected to evaluate the impact of the first year of the Sesame Street television series. The main concern of Sesame Street was in teaching preschool related skills to children in the 3-5 years age range, with special emphasis on reaching 4-year-old children identified as disadvantaged. The format of the show was designed to hold young children’s attention through action oriented, short duration presentations teaching specific preschool cognitive skills and some social skills. Each show was scheduled to take one hours and it involved much repetition of concepts within and across shows. The evaluation was conducted by Education Testing Services (ETS). Five sites were of interest to the producers of the program (the children’s Television Workshop: Three to five year-olds disadvantaged from inner city areas in various parts of the country, four-year-old advantaged suburban children, advantaged rural children, disadvantaged rural children, and disadvantaged Spanish speaking children. Research question(s) and hypotheses The purpose of this study was to analyze data obtained from ETS researchers looking at the cognitive ability of children watching Sesame Street which is a children’s education show. We tested the null hypothesis that there is NO statistical significant difference in post-body perfectionism scores of preschool children from five locations (disadvantaged from inner city areas, advantaged suburban children, advantaged rural children, disadvantaged rural children, and disadvantaged Spanish speaking children). The alternative hypothesis for this study was there is a statistical significant difference in post-body perfectionism scores of preschool children from five locations (disadvantaged from inner city areas, advantaged suburban children, advantaged rural children, disadvantaged rural children, and disadvantaged Spanish speaking children). The research questions were as follows: Do the different samples sites present statistically significant mean differences on post-body perfectionism scores of preschool children? If so, which sample site differs? Variables The dependent variable included in this research study was post-body perfectionism scores that were continuous in nature. The subject was to give any value for post-body perfectionism scores. The sample sites were the independent variables that were categorical in nature. A child was to belong to any of the five demographic locations. Children were selected from five demographic locations coded as follows: disadvantaged inner city; advantaged suburban; and advantaged and disadvantaged rural settings. Disadvantaged Spanish speaking children were also included in this study. The design for this analysis is unbalanced based on the inclusion of disadvantaged Spanish speaking children sampling population (n=18) that is compared to the other four groups: disadvantaged inner city (n=58), advantaged suburban(n=55), advantaged and disadvantaged rural settings (n=64) and (n=43) respectively. Disadvantaged Spanish speaking children sampling population sample site could be excluded since it meets Meyers, Gamst, and Guiarino (2006) criteria that a design is unbalanced when the larger cell number is at least 3 times greater than the smallest cell number. In this case the largest cell number is 58 compared with 18. The rest of the samples apart from Spanish speaking children sampling population qualify for Power analysis of 0.80 since they have n>20 (Myoung, 2010). Data analysis methods A one-way between subjects ANOVA was conducted to test if there was a statistically significant difference among children’s perfectionism scores by site. To conduct the one-way ANOVA the following assumptions must be met: The dependent variables must have independent observations, the dependent variables have univariate normality, and there should be a homogeneous variance-covariance matrix (Sheskin, 2003). The researchers tested the data for the assumptions. Using Levene statistics, with null hypothesis that there is no homogeneity of variance against the alternative hypothesis that there is homogeneity of variance, Levene test indicated a significant results with F= 19.832, and p Read More

The test was to check if there was a significant difference in pre self-esteem of runners by gender (Male and female). This statistical test is appropriate when one has separate samples under the same treatment condition (Heiman, 2013). In this case the treatment was running. The population mean and standard deviation are normally not known. The key assumptions for the test are the sample populations under study must be independent, the scores in each population must be normally distributed, and the two populations must have equal variance (Sheskin, 2003).

From the descriptive statistics above, the difference in the means indicate that these samples were independent of each other. The researcher screened the data before the actual analysis by independent sample t-test. Frequency Table 1 above indicated that there were no missing values. The box plot also indicated that there were no outliers that could reflect data entry errors in the study. The skewness and kurtosis values were within the acceptable range of -1.0 to 1.0 (Meyers et al. (2006).

Results An independent sample t-test of pre self-esteem by gender was conducted. Male (n=17, M=29.24) and female (n=13, M=30.46) had standard deviations SD=6.34 and SD=5.7 respectively. This indicates that the two were not further spread from their means since their standard deviations were within acceptable range. The means of the two genders indicated that female gender scored higher than male on the pre self-esteem scale (see Table 3 below). Table 3 Descriptive Statistics for Pre-self steem Score by Gender Gender n M (SE) SD Male 17 29.24 6.34 Female 13 30.46 5.60 Homogeneity of variance on the pre self-esteem data collected by gender was assessed using Levene’ test of equality of variance.

Levene test indicated non-significant results, F=0.494, p>0.05. This indicated that the assumption on homogeneity of variance was met. The samples varied in testing for normal distribution as the pre self-esteem for male had a small negative skewness while pre self-esteem for female had zero skewness. According to Meyers, et al. (2006) criteria, skewness was within tolerance level of between -1.0 and 1.0 but kurtosis was not. The data was however adopted for further analysis. Male Skewness -.

470 Kurtosis -.805 Female Skewness 0.000 Kurtosis -1.052 Independent sample t-test results showed that there was no statistically significant difference in average pre self-esteem score by sex, t 28=,-.548 p > .05. This means that the researchers failed to reject the null hypothesis and conclude that there is no statistically significant difference in runners’ pre self-esteem score by gender. According to Cohen (1992), the effect size of the study is medium since d=0.793. This is supported by the standard deviation units of genders.

Male (SD=6.34) and female (SD=5.60) indicate that the values were not so much spread and thus could have a medium effect. References Cohen, J. (1992). Quantitative methods in psychology. Psychological Bulletin, 112(1), 155-159. Heiman, G. W. (2013). Basic Statistics for the behavioral sciences. Belmont CA: Cengage learning. Meyers, L. S., Gamst, G., & Guarino, A. J. (2006). Applied multivariate research: Design and interpretation. Thousand Oaks, CA: Sage. Sheskin, D. J. (2003). Handbook of parametric and nonparametric statistical procedures ed.

Boca Raton, Fla: Chapman & Hall/CRC Press. Dependent samples t-test Brief description of the study This study used the data file named “A study of the effectiveness of a 6-week running weight loss program”. This study involved 30 individuals who volunteered to participate in a 6-week weight loss running program. The researcher runner met at a local running specialty shoe store twice during the week (Monday to Friday) as a group over 6-week period and once on weekends at a local park (either Saturday or Sunday).

Those who entered the program were interested in losing weight. A group plan was created to follow over the 6 week period by the researcher/trainer.

Read More