The effect of Media in Information Technology and Job creation

CHAPTER FOUR

DATA ANALYSIS, PRESENTATION AND INTERPRETATION OF RESULTS

4.1 Introduction

This chapter encompasses the analysis of the data collected from the field on the research topic: the effect of social network Platforms on Employment Generation through questionnaires. The questionnaire was constructed with four sections, all of which generated the data for the analysis. The Statistical Package for Social Sciences (SPSS) was used to analyze the data after which the results will be interpreted. The questionnaire has a total of 35 questions and four sections of which the last section was exclusively for HR personnel and business owners.

A total of 100 questionnaires were administered during the survey stage of this research project. They were administered via the internet through the use of survey monkey, a web-based survey medium. Of the 100 questionnaires administered, 100 questionnaires were returned indicating a
response of 100%.

4.2 Presentation of Data

4.2.1 Socio-Economic Distribution Table Of Respondents

TABLE 4.1

Socio-Economic Factor Percentage Socio-Economic Factor
Percentage Employment Status Gender
Employed 87.0 Female 57.0 Unemployed 13.0 Male 43.0 Total 100.0
Total 100.0
Age Group Educational Attainment
Below 25 29.0 Hnd/Ond 9.0 25-30 25.0 BSc/Ba 52.0 36-40 5.0
41-45 12.0 MSc 38.0 PhD 1.0 Above 45 11.0 Total 100.0 Total 100.0
Source; field survey, 2014.

The table above shows the percentage distribution of the respondents by their socio-economic characteristics. The first question asked was about employment. This is because the basis of the research is on employment generation. Out of the valid 100 respondents, 87% representing 87 respondents were employed while 13% representing 13 respondents were unemployed.

We also grouped the age brackets into six different parts in order to see if the use of social media is limited to only one age group or if it cuts across all the ages. From the frequency table above, we see that of the 100 valid responses, 29% representing 29 respondents are those below the ages of 25, 25% representing 25 respondents are those between the ages of 25 and 30, 18% representing 18 respondents are those between the ages 31 and 35, 5% representing 5 respondents are those between the ages of 36 and 40, 12% representing 12 respondents are those between the ages of 41 and 45 and 11% representing 11 respondents are those above 45 years.

 A question on gender was also asked to be sure that the questionnaire cuts across both gender and is representative. From the valid 100 responses, 57% were female responses while 43% were male responses, showing that the majority of the respondents were female. The research also cuts across people with different educational backgrounds out of the total 100 respondents, 9% are HND/OND holders, 52% are BSc/Ba degree holders, 38% are MSc degree holders and 1% PhD degree holders.

4.3 Analysis and Interpretation of Results

TABLE 4.2:

Factors Affecting Employment Opportunities Employment Opportunities Percentage Professional Use Percentage

YES 69.0 YES 88.0 NO 31.0 NO 12.0 Total 100.0 Total 100.0 Parents Social Status Firms Use of SN For Recruiting Upper Class 26.0 YES 67.0 Middle Class 67.0 NO 32.0 Lower Class 7.0 Total 99.0 Total 100.0 Missing System 1.0 Total 100.0 Social Presence SN Is Better Than Trad Methods

YES 98.0 YES 75.0 NO 2.0 NO 25.0 Total 100.0 Total 100.0
Professional Use SN In Connecting Both Parties

YES 88.0 EFFECTIVE 83.0 NO 12.0 INEFFECTIVE 17.0 Total 100.0
Total 100.0

Source; field survey, 2014

From the table above, we see that majority of the people who filled the questionnaire 69% have gotten more employment opportunities since they started using the Social network sites as against those who have not gotten who represent 31% of the sample population also showing that these sites help improve your chances of getting a job. Out of all the 100 respondents, 26% had parents who were in the upper class of society, 67% of them had parents who were in the middle class of society and 7% were members of the lower class of society, showing that apart from the Social Network platforms, other factors such as the parents status could have been responsible for the high level of employment recorded. Of the 100 respondents, 98% have social network presence while 2% do not. This shows that the sample population was well captured has the research hopes to show that being on a social network increases your chances of getting a job. Of the 100 respondents, and 98 active social media users, 88 of them use it for professional purposes.

This is to help show that the use of social network is in line with the topic in hand. Out of the 100 respondents, 99 answered this question out of which 67% said the firms they work in use Social Networks for recruitment purposes while 32% said they do not. These figures show that truly firms are moving from the traditional methods of recruiting to social recruiting. From a priori knowledge on the theory of social exchange, we know that individuals would always go for recruitment methods that have the best cost-benefit ratio and with social recruiting, there is minimal cost with countless benefits so getting a result in which 75% believe social recruiting to be better than traditional methods while only 25% believe otherwise is not surprising. The main aim of social recruiting is its ability to bring together both parties of the labour market more effectively than the traditional means of recruiting. From the results gotten from this survey we see that indeed social Networks are effective in connecting employers and those in search of jobs. This is evident from the 83% who say that it is effective.

TABLE 4.3: Factors Affecting Employment Generation.
Level Of Productivity With SR Percentage of New Recruits Via SN
Percentage
INCREASED 50.0 HIGH 7.0 REDUCED 4.0 MEDIUM 24.0 REMAINED
UNCHANGED 46.0 LOW 13.0 Total 44.0 Total 100.0 Missing System
56.0 Increase in Sales Job Openings Yes 32.0 GREATER 26.0 No 12.0 LESS 2.0 Total 44.0 UNCHANGED 15.0 Missing System 56.0 Total 43.0 Total 100.0 Missing System 57.0 Source; field survey, 2014

Social recruiting in creating more employment can only be possible if there has been increase in productivity either via the quality of the new employees or increase in sales from advertising on these sites. From the responses gotten, we see the people in the category of those who believe that their output increased 50% is higher than those who believed that it reduced 4% and those who believe it
remained unchanged 46% which shows that indeed SR helps in creation of jobs. Of the 45 respondents eligible to fill this part of the questionnaire, 44 answered this question and 32 out of these respondents believe that they have had an increase in sales that can be attributed to their social network presence which is the premise for increase in employment.

Also in a bid to determine if firms actually do recruit via social media, we asked them this question and we saw that 7 of them have a high percentage of SR while the majority i.e. 24 have a medium percentage of SR and 13 of them have low percentage of Social recruits. These figures are quite impressive seeing as how Social Recruiting is a newly introduced concept in Nigeria so the fact that it is being used to this extent is good news. Lastly, the table above shows the actual relationship between social recruiting and employment generation. 26 of the 43 respondents said that there have been more openings, 2 said less and 15 said it remained unchanged this shows that indeed, there lies a positive relationship between social network presence and employment generation.

4.4. Regression Analysis

After frequency analysis and has been carried out, it is imperative that further steps be taken to test the hypothesis designed for the study. In the first chapter of the study, hypotheses were formulated, this section examined the hypothesis to prove if they are significant or not. The statistical technique employed here is the regression analysis.

By definition, regression is a technique that provides mathematical modeling and analysis of numerical data consisting of a dependent variable and of one or more independent variables. It is a statistical methodology that utilizes the relationship between 2 or more variables. When it involves one dependent and one independent variable, it is linear regression, on the other hand, when it involves one dependent variable and one or more independent variable, it is a multiple regression.

4.4.1 Logit Regression Analysis of Social Network Presence and Employment
Opportunities

Logistic regression allows you to access how well your set of predictor variables predicts or explains your categorical dependent variable. It gives you an indication of the adequacy of your model (set of predictor variables) by assessing ‘goodness of fit’. It provides an indication of the relative importance of each predictor
variable or the interaction among your predictor variables.

Table 4.4: Dependant Variable: Employment Generation Coefficients B T-Stat Constant Control Variables

SP 2.801* (.757) Yes 0.812** (.057) 2.433 Yes 4.765 EA .365** (.395) 5.356 PSS .406* (.477) 2.395
49
SA 0.454* (.925) 3.008
F-Stat 4.282* R 0.487 R
2
26.3% Omnibus 14.278* df6
*Significant at 95%
**Significant at 99% Control variables include Age Group, Religion and Gender. Which when tested for all came back significant
From the table above, there appears to be a positive relationship between employment opportunities (dependent variable) and age group, gender, religion, educational attainment, parents’ social status, social network presence and social network activeness (independent variables). A unit increase in Social Network Presence will result in a corresponding 0.812 increase in employment opportunities. A unit increase in Educational Attainment (going a level higher) will bring about a corresponding 0.365 increase in Employment Opportunities. A unit increase in Parents Social Status will bring about a corresponding 0.406 increase in Employment Opportunities. A unit increase in Social Network activeness, will lead to a corresponding 0.454 increase in Employment opportunities. The t statistic is a measure of the significance of the independent variable on the dependent variable. The absolute T value of for all the variables are greater than 2 we therefore reject H0 and conclude that Social network presence has an impact on
employment Opportunities 99% and 95% significance levels. The F statistics is used to project the joint statistical significance of all independent variables on the dependent variable. The value of the F statistic above is highly statistically significant at 5% level of significance. Therefore, the employment status is significantly related to social network presence.

The R value is a measure of the strength of the relationship between the dependent variable and the independent variable. R value of 0.487 indicates that there is a strong relationship between Social Network Presence and Employment Generation. The R2 on the other hand shows the degree of variation of the dependent variable explained by the model. 26.3% of the variation in employment generation is explained by the model. Thus, the model fits the population. The Omnibus Test of Model Coefficient gives us an overall indicator of how the model acts even when none of the independent variables are added. It is a test of goodness of fit. From the test carried out, we have a chi-square value is significant at 14.278 at 6 degree of freedom.

4.4.2 Hosmer and Lemeshow Test

The Hosmer and Lemeshow test is also another test under the logit analysis. It is the most reliable test of model fit. It is however interpreted differently. In this case, the model is of great fit when the level of significance is greater than 0.05.
TABLE 4:

Hosmer and Lemeshow Test

l Chi-square df Sig.
1 3.982 8 .859
In this model, the chi-square value of Hosmer and Lemeshow is 3.982 with a 0.859 level of significance which shows that the model has a very good fit.

4.4.3 OLS

Regression Analysis of Social Network Presence and Employment
Opportunities
TABLE 4.6
Dependent Variable: EMPLOYMENT OPPURTUNITIES (1) (2) Var iab B
T-Variables B T-Stat
51
les Stat
(Constant) 1.119** (.549) 2.036 (Constant) 0.472 2.442 Contr ol Var
iables Yes Yes Control Variables Yes Yes EA .452** (.071) 4.996 EA
0.384** (0.73) 2.453 PSS .343** (.084) 6.830 PSS 0.294** (.882)
1.813 SP .755** (.360) 5.098 SP NO NO SN A .50 3.230 SNA NO NO
52
8*** (.157)
F-Stat 3.315** F-Stat 1.850* R 0.449 R 0.299 R
2
20.1% R
2
9%
*Significant at 90%
**Significant at 95% ***Significant at 99% Control variables include Age Group, Religion and Gender. Which when tested for all came back significant. From the table above, in the section labelled (1) there appears to be a positive relationship between employment opportunities (dependent variable) and age group, gender, religion, educational attainment, parents social status, social network presence and social network activeness (independent variables). A unit increase in Educational Attainment (going a level higher) will bring about a corresponding 0.452 increase in Employment Opportunities. A unit increase in Parents Social Status will bring about a corresponding 0.343 increase in Employment Opportunities. A unit increase in Social Network activeness, will lead to a corresponding 0.508 increase in Employment opportunities. A unit increase in Social Network Presence will result in a corresponding 0.755 increase in employment opportunities. The statistic is a measure of the significance of the independent variable on the dependent variable.

The absolute T value of for all the variables are greater than 2 we therefore reject H0 and conclude that Social network presence has an impact on
employment opportunities, 99% and 95% significance levels.

The F statistics is used to project the joint statistical significance of all independent variables on the dependent variable. The value of the F statistic above is highly statistically significant at 5% level of significance. Therefore, the employment status is significantly related to social network presence. The R value is a measure of the strength of the relationship between the dependent variable and the independent variable. R value of 0.449 indicates that there is a strong relationship between social network presence and employment generation.

The R2 on the other hand shows the degree of variation of the dependent variable explained by the model. 20.1% of the variation in employment generation is explained by the model. Thus, the model fits the population. A second regression was carried out to test the actual effect of social network presence and activeness on employment opportunities. In this regression, these two variables
were removed. The results are represented in the part labelled (2). From the table, we see that there is still a positive relationship between employment opportunities
(dependent variable) and age group, gender, religion, educational attainment, parents’ social status (independent variables). A unit increase in Educational Attainment (going a level higher) will bring about a corresponding 0.384 increase in Employment Opportunities. A unit increase in Parents Social Status will bring about a corresponding 0.294 increase in Employment Opportunities. The t statistic is a measure of the significance of the independent variable on the dependent variable. It is only Educational Attainment that has a significant relationship with employment opportunities. From this, we see that in this sample population. The other variables such as Parents Social Status, Age group, Gender and religion do not have a significant relationship with employment opportunities when social network presence and activeness is absent.

The F statistics is used to project the joint statistical significance of all independent variables on the dependent variable. The value of the F statistic above is significant at 10% level of significance. The low level of significance shows that there is a low relationship between the independent variables and the dependent variable. The R value is a measure of the strength of the relationship between the dependent variable and the independent variable. R value of 0.299 indicates that there is a weak relationship between social network presence and employment generation. The R2 on the other hand shows the degree of variation of the dependent variable explained by the model. 9% of the variation in employment generation is explained by the model as against the 20.1% variation in the previous model

4.5.3 Regression Analysis of Employment Generation and Increased Productivity

 TABLE 4.7Dependent Variable: EMPLOYMENT GENERATION (1) (2) Variable B
T-Stat Variable B T-Stat
(Constant) 1.078** (.715) 2.508 (Constant) 0.835** (0.425) 2.254 IS
.668** (.349) 4.915 IS NO NO
55
IP .885** (.149) 3.566 IP NO NO FS .539** (.234) 2.024 FS 0.458**
(0.226) 2.141 EQ .695** (.177) 3.028 EQ NO NO *
Significant at 90% **Significant at 95% ***Significant at 99%
F-Stat 1.506** F-Stat 1.312* R 0.383 R 0.218 R
2
14.7% R
2
7% Correlation 0.719***
From the table above, in the section labelled (1), there is a positive relationship between employment generation (dependent variable) and increased sales, increased productivity, size of firm and quality of employees with SR (independent variables). A unit increase in Sales (going a level higher) will bring about a corresponding 0.668 increase in Employment Generation. A unit increase in Productivity will bring about a corresponding 0.885 increase in Employment Generation. A unit increase in Firm Size, will lead to a corresponding 0.539 increase in Employment Generation. A unit increase in Employee Quality via SR will result in a corresponding 0.695 increase in employment opportunities. The t-statistic is a measure of the significance of the independent variable on the dependent variable. The absolute T value of for all the variables are greater than 2 we therefore reject H0 and conclude that Increased Productivity has an impact on
employment Generation at a 95% level of Significance. The F statistics is used to project the joint statistical significance of all independent variables on the dependent variable. The value of the F statistic above is highly statistically significant at 5% level of significance. Therefore, the Employment Generation is Significantly Related to Increased Productivity via SN. The R value is a measure of the strength of the relationship between the dependent variable and the independent variable. R value of 0.383 indicates that there is a relationship between Employment Generation and Productivity.

The R2 on the other hand shows the degree of variation of the dependent variable explained by the model. 14.7% of the variation in employment generation is explained by the model. Thus, the model fits the population. A test of reliability was also carried out on this hypothesis to test the accuracy of the test. For this, a Pearson-Correlation was carried out in which came out significant at 99% showing that the regression was indeed accurate. Also, for this hypothesis, all variables related to social networks were removed in order to see the impact social networks have on employment generation in this sample population. The results of this regression are represented in the section (2) of the table above. There is a positive relationship between employment generation (dependent variable) and firm size (independent variable).

A unit increase in Firm Size, will lead to a corresponding 0.458 increase in Employment Generation. The t-statistic is a measure of the significance of the independent variable on the dependent variable. The absolute T value of for firm size is greater than 2 we therefore reject H0 and conclude that firm size has an impact on employment Generation at a 95% level of Significance. The F statistics is used to project the joint statistical significance of all independent variables on the dependent variable. The value of the F statistic above is weakly statistically significant at 10% level of significance. Therefore, the Employment Generation is significantly related to firm size although in this case, the relationship is not very strong The R value is a measure of the strength of the relationship between the dependent variable and the independent variable. R value of 0.218 indicates that there is a relationship between Employment Generation and firm size. The R
2
on the other hand shows the degree of variation of the dependent variable explained by the model. 7% of the variation in employment generation is explained by the model. This is quite low when compared with the 14.7% variation in the previous model.
4.4.4 CHI-SQUARE ANALYSIS OF EFFECTIVENESS OF SOCIAL
NETWORKS TABLE 4.8
SN IN CONNECTING BOTH PARTIES
Obser ved N Expected N EFFECTIVE 83 50.0 INEFFECTIVE 17 50.0
Total 100

TABLE 4.9 Test Statistics

SN IN CONNECTING BOTH PARTIES Chi-Square(a) 43.560 df 1
Asymp. Sig. .000

At degree of freedom 1 and at 1% level of significance, the table value of the chi-square. Since the calculated value is greater than the tabulated value; we accept the alternative hypothesis and conclude that social networks are effective in connecting employers and people looking for jobs together.

4.5 Conclusion

This chapter included the data presentation and analysis. Each of the three hypotheses were tested and found to be true. For the first hypothesis, a Logit regression was first carried out to test the relationship between the variables. From this, we saw that there was indeed a significant relationship between the variables. Next, I carried out two OLS regressions. In the first regression, I used all of the independent variables which showed that there was a significant relationship between these variables and employment opportunity. To test the magnitude of this effect, a second OLS regression was carried out in which I excluded social network presence and social network activeness. The new result showed a significant relationship only between educational attainment and employment opportunity. This showed that in this study population, all of the other factors were not significant when social network presence was absent For the second hypothesis, I also carried out two OLS regressions to show the relationship that exists between increased productivity via social network presence and employment generation.

The first regression included the all of the independent variables while in the second regression I excluded all variables related to social networks. The results got showed a weaker relationship than that of the model with social network presence. For the third hypothesis, I carried out a chi-square analysis which came back significant. From all the tests carried out, It was found out that social network presence increases job opportunities for the individuals, creates more jobs openings in firms and is effective in bringing employers and those in search of jobs together. Read chapter here